From 65d0fd389c61fc3a98d150be789c33da39e36f6a Mon Sep 17 00:00:00 2001 From: minixalpha Date: Fri, 30 Oct 2026 08:31:51 -0700 Subject: [PATCH 01/15] feat(benchmarks): add the official deepseek harness adapter --- benchmarks/harbor/README.md | 27 +++ benchmarks/harbor/README.zh-CN.md | 23 +++ benchmarks/harbor/scripts/run_comparison.py | 5 + .../harbor/src/harbor_adapter/__init__.py | 3 +- .../harbor/src/harbor_adapter/__main__.py | 46 ++++- .../harbor/src/harbor_adapter/_dsh_client.py | 53 +++++ .../harbor/src/harbor_adapter/deepseek.py | 188 ++++++++++++++++++ benchmarks/harbor/tests/test_deepseek.py | 184 +++++++++++++++++ 8 files changed, 526 insertions(+), 3 deletions(-) create mode 100644 benchmarks/harbor/src/harbor_adapter/_dsh_client.py create mode 100644 benchmarks/harbor/src/harbor_adapter/deepseek.py create mode 100644 benchmarks/harbor/tests/test_deepseek.py diff --git a/benchmarks/harbor/README.md b/benchmarks/harbor/README.md index eb667fa..080df0d 100644 --- a/benchmarks/harbor/README.md +++ b/benchmarks/harbor/README.md @@ -148,6 +148,33 @@ preserves all 34 attempts, the supervision amendment, two unscored legacy timeouts, public-reference exposure, and partial accounting. It does not claim a complete, uncontaminated 34-task pass rate. +### DeepSeek Harness baselines + +`harbor_adapter:DeepSeekHarnessAgent` runs the upstream Python SDK's +`sdk-minimal` profile, as recommended by DeepSeek Harness's `BENCHMARK.md`. +Register `distribution_path`, `distribution_sha256`, and optional +`reasoning_effort` (default `high`) in the agent kwargs. The distribution is an +offline `wheels/` directory containing the SDK, native runtime, and all Python +dependencies, plus `distribution.json` with `version`, `source_commit`, +`source_archive_sha256`, `sdk_wheel`, and `runtime_wheel`. The digest covers every +file and permission; the standard workflow snapshots and rechecks the bytes. + +Build the runtime and SDK from the selected upstream checkout using its documented +Python contributor workflow. Task installation uses Python 3.13 and the frozen +wheelhouse. Installation-only checks initialize the actual SDK profile and model +without sending a task prompt, and run the existing verifier dependency preflight. +Model work requires `DEEPSEEK_API_KEY` and uses the official Messages endpoint. +Each task receives a fresh Harness home and session, with upstream prompts, +tools, compaction, retries, and generation defaults. Native logs remain in +`agent/dsh/`; SDK events, completion status, and diagnostics remain beside it. + +Configurations must set `verifier.kwargs.require_agent_quiescence=true`. +The shared Linux supervisor confirms cancellation of the agent process tree +before verification. A non-completed SDK finish reason is an execution error. +Reported tokens include native assistant and compaction usage; missing records +and interrupted runs remain partial, and cost remains unknown. This baseline +does not impose Pi's generation settings or Nano's turn and work budgets. + ### Cache original task images If Docker image pulls repeatedly stall, the optional cache utility downloads the diff --git a/benchmarks/harbor/README.zh-CN.md b/benchmarks/harbor/README.zh-CN.md index d7b3d9c..1b249df 100644 --- a/benchmarks/harbor/README.zh-CN.md +++ b/benchmarks/harbor/README.zh-CN.md @@ -116,6 +116,29 @@ Docker 空间受限时,可以通过 `batch_images` 计划按批恢复已注册 保留了全部 34 次尝试、进程托管修正、两次不评分的旧超时、公开参考材料接触和 不完整用量统计,不宣称完整且无答案接触的 34 题通过率。 +### DeepSeek Harness baseline + +`harbor_adapter:DeepSeekHarnessAgent` 使用上游 Python SDK 的 `sdk-minimal` +配置运行,这是 DeepSeek Harness 的 `BENCHMARK.md` 推荐的基准配置。 +在 agent kwargs 中注册 `distribution_path`、`distribution_sha256` 和可选的 +`reasoning_effort`(默认为 `high`)。分发目录包含离线 `wheels/` 目录,内有 +SDK、原生运行时以及所有 Python 依赖;另有 `distribution.json`,记录 +`version`、`source_commit`、`source_archive_sha256`、`sdk_wheel` 和 +`runtime_wheel`。摘要覆盖每个文件及其权限;标准流程会保存快照并重新校验。 + +按照选定上游检出的 Python 贡献者文档构建运行时和 SDK。任务安装使用 +Python 3.13 和冻结的 wheel 集合。仅安装检查会初始化真实 SDK 配置和模型, +但不发送任务提示,并执行已有的验证器依赖预检。模型执行需要 +`DEEPSEEK_API_KEY`,使用官方 Messages 端点。每道题都使用全新的 Harness +目录与会话,保留上游提示词、工具、上下文压缩、重试及生成默认值。原生日志 +保存在 `agent/dsh/`,SDK 事件、完成状态和诊断信息保存在旁边。 + +配置必须设置 `verifier.kwargs.require_agent_quiescence=true`。 +共享的 Linux 进程监督器在验证前确认 agent 进程树已被取消。SDK 返回非 +`completed` 的结束原因时记为执行错误。报告的 token 包含原生助手消息和 +压缩请求用量;缺失记录和中断运行仍标记为部分统计,费用保持未知。 +此 baseline 不强制套用 Pi 的生成设置或 Nano 的轮数和工作时间预算。 + ### 缓存原始任务镜像 如果 Docker 拉取镜像反复停滞,可用缓存工具从同一 Docker Hub 公共仓库分段并行下载: diff --git a/benchmarks/harbor/scripts/run_comparison.py b/benchmarks/harbor/scripts/run_comparison.py index 57891f7..9904b37 100644 --- a/benchmarks/harbor/scripts/run_comparison.py +++ b/benchmarks/harbor/scripts/run_comparison.py @@ -148,6 +148,11 @@ def validate_plan(root, plan_path, plan): from harbor_adapter.pi import distribution_digest, distribution_files if distribution_digest(distribution_files(Path(arm["distribution_path"]))) != arm["distribution_sha256"]: raise ValueError("Registered Pi distribution changed") + elif arm.get("agent") == "deepseek-harness": + from harbor_adapter.deepseek import distribution_files + from harbor_adapter.pi import distribution_digest + if distribution_digest(distribution_files(Path(arm["distribution_path"]))) != arm["distribution_sha256"]: + raise ValueError("Registered DeepSeek distribution changed") else: expected = plan_path.parent / arm["prompt_file"] if digest(expected) != arm["prompt_sha256"]: diff --git a/benchmarks/harbor/src/harbor_adapter/__init__.py b/benchmarks/harbor/src/harbor_adapter/__init__.py index b27118f..63c7fa8 100644 --- a/benchmarks/harbor/src/harbor_adapter/__init__.py +++ b/benchmarks/harbor/src/harbor_adapter/__init__.py @@ -1,7 +1,8 @@ """Public import for the repository-local nanoPyCodeAgent Harbor adapter.""" from .adapter import NanoPyCodeAgent +from .deepseek import DeepSeekHarnessAgent from .pi import PiCodeAgent from .verifier import RetryingVerifier -__all__ = ["NanoPyCodeAgent", "PiCodeAgent", "RetryingVerifier"] +__all__ = ["NanoPyCodeAgent", "PiCodeAgent", "DeepSeekHarnessAgent", "RetryingVerifier"] diff --git a/benchmarks/harbor/src/harbor_adapter/__main__.py b/benchmarks/harbor/src/harbor_adapter/__main__.py index 57baead..40ae1b7 100644 --- a/benchmarks/harbor/src/harbor_adapter/__main__.py +++ b/benchmarks/harbor/src/harbor_adapter/__main__.py @@ -16,10 +16,12 @@ from harbor.models.job.config import JobConfig from .bootstrap import AptCache, default_cache_dir +from .deepseek import distribution_files as deepseek_distribution_files from .pi import distribution_digest, distribution_files, read_model_config _NANO = "harbor_adapter:NanoPyCodeAgent" _PI = "harbor_adapter:PiCodeAgent" +_DEEPSEEK = "harbor_adapter:DeepSeekHarnessAgent" def standard_config(raw: dict) -> JobConfig: @@ -56,11 +58,13 @@ def standard_config(raw: dict) -> JobConfig: raise ValueError("Select at least one agent and one trial attempt") for agent in config.agents: identifiers = [value for value in (agent.import_path, agent.name) if value] - if not identifiers or len(set(identifiers)) != 1 or identifiers[0] not in {_NANO, _PI}: + if not identifiers or len(set(identifiers)) != 1 or identifiers[0] not in {_NANO, _PI, _DEEPSEEK}: raise ValueError("The standard workflow requires a repository benchmark adapter") allowed = ({"max_turns", "max_tokens", "time_budget_seconds"} if identifiers[0] == _NANO else { "distribution_path", "distribution_sha256", "model_config_path", "thinking", + } if identifiers[0] == _PI else { + "distribution_path", "distribution_sha256", "reasoning_effort", }) if agent.env or set(agent.kwargs) - allowed: raise ValueError( @@ -72,11 +76,45 @@ def standard_config(raw: dict) -> JobConfig: if not required <= agent.kwargs.keys(): raise ValueError("Pi requires a hash-pinned distribution and public model config") read_model_config(Path(agent.kwargs["model_config_path"])) + if identifiers[0] == _DEEPSEEK: + if not {"distribution_path", "distribution_sha256"} <= agent.kwargs.keys(): + raise ValueError("DeepSeek requires a hash-pinned distribution") + if config.verifier.kwargs.get("require_agent_quiescence") is not True: + raise ValueError("DeepSeek requires confirmed agent quiescence before grading") + if agent.kwargs.get("reasoning_effort", "high") not in {"off", "low", "high", "max"}: + raise ValueError("Unsupported DeepSeek reasoning effort") if not agent.model_name: raise ValueError("Each agent requires an explicit model_name") return config +def snapshot_deepseek(agent, cache: Path, record: Path) -> dict: + """Publish a verified, immutable offline wheelhouse for all task installs.""" + original = Path(agent.kwargs["distribution_path"]).expanduser().resolve() + files = deepseek_distribution_files(original) + digest = distribution_digest(files) + if digest != agent.kwargs["distribution_sha256"]: + raise ValueError("DeepSeek distribution differs from its registered SHA256") + destination = cache / "deepseek-distributions" / digest + if not destination.exists(): + destination.parent.mkdir(parents=True, exist_ok=True) + with tempfile.TemporaryDirectory(prefix=digest + ".tmp-", dir=destination.parent) as temporary: + staged = Path(temporary) / "distribution" + shutil.copytree(original, staged) + if deepseek_distribution_files(staged) != files: + raise ValueError("DeepSeek distribution changed while snapshotting") + try: + staged.rename(destination) + except OSError as error: + if error.errno not in {errno.EEXIST, errno.ENOTEMPTY}: + raise + if deepseek_distribution_files(destination) != files: + raise ValueError("Cached DeepSeek distribution is corrupt") + agent.kwargs["distribution_path"] = str(destination) + return {"distribution_sha256": digest, "files": files, + "provenance": json.loads((destination / "distribution.json").read_text())} + + def snapshot_pi(agent, cache: Path, record: Path) -> dict: """Share one content-addressed binary snapshot across the full task pool.""" original = Path(agent.kwargs["distribution_path"]).expanduser().resolve() @@ -174,10 +212,12 @@ def run(args) -> int: for path in (source / "nanopycodeagent").rglob("*.py"): if archive.read(path.relative_to(source).as_posix()) != path.read_bytes(): raise ValueError("Source changed during preparation; prepare a new job") - pi_inputs = [] + pi_inputs, deepseek_inputs = [], [] for agent in config.agents: if (agent.import_path or agent.name) == _PI: pi_inputs.append(snapshot_pi(agent, cache, record)) + elif (agent.import_path or agent.name) == _DEEPSEEK: + deepseek_inputs.append(snapshot_deepseek(agent, cache, record)) else: agent.kwargs["wheel_path"] = str(wheel) agent.kwargs["bootstrap_cache"] = str(cache) @@ -194,6 +234,8 @@ def run(args) -> int: } if pi_inputs: manifest["pi_inputs"] = pi_inputs + if deepseek_inputs: + manifest["deepseek_inputs"] = deepseek_inputs (record / "manifest.json").write_text(json.dumps(manifest, indent=2) + "\n") (record / "working-tree.patch").write_bytes(subprocess.check_output(["git", "diff", "HEAD"])) print(f"Prepared {record}", flush=True) diff --git a/benchmarks/harbor/src/harbor_adapter/_dsh_client.py b/benchmarks/harbor/src/harbor_adapter/_dsh_client.py new file mode 100644 index 0000000..403ba59 --- /dev/null +++ b/benchmarks/harbor/src/harbor_adapter/_dsh_client.py @@ -0,0 +1,53 @@ +"""Container-side driver for the upstream DeepSeek Harness Python SDK.""" + +import argparse +from dataclasses import asdict +import json +from pathlib import Path + + +def main(): + from deepseek_harness import DeepSeekHarness + + parser = argparse.ArgumentParser(description=__doc__) + parser.add_argument("--home", required=True) + parser.add_argument("--model", required=True) + parser.add_argument("--reasoning-effort", default="high") + parser.add_argument("--preflight", action="store_true") + parser.add_argument("--instruction") + args = parser.parse_args() + home = Path(args.home) + home.mkdir(parents=True, exist_ok=True) + with DeepSeekHarness( + provider="deepseek-official", model=args.model, + reasoning_effort=args.reasoning_effort, profile="sdk-minimal", + cwd=str(Path.cwd()), dsh_home=str(home), + ) as harness: + # Initialization resolves the real profile/provider/model without a prompt. + if args.preflight: + print(json.dumps({"initialized": True, "model": args.model})) + return 0 + if args.instruction is None: + parser.error("--instruction is required for model work") + with (home.parent / "dsh-events.jsonl").open("w") as events: + def notification(event): + # Durable session events contain usage and tool results. Omit + # streaming projections, whose text duplicates the native log. + if event.method == "session.event": + events.write(json.dumps(asdict(event)) + "\n") + events.flush() + + result = harness.run( + Path(args.instruction).read_text(), session_id="benchmark", + on_notification=notification, + ) + (home.parent / "dsh-result.json").write_text(json.dumps({ + "session_id": result.session_id, + "finish_reason": result.finish_reason, + "final_response": result.final_response, + }) + "\n") + return 0 if result.finish_reason == "completed" else 1 + + +if __name__ == "__main__": + raise SystemExit(main()) diff --git a/benchmarks/harbor/src/harbor_adapter/deepseek.py b/benchmarks/harbor/src/harbor_adapter/deepseek.py new file mode 100644 index 0000000..0746d9d --- /dev/null +++ b/benchmarks/harbor/src/harbor_adapter/deepseek.py @@ -0,0 +1,188 @@ +"""Run the upstream DeepSeek benchmark profile from hash-pinned local wheels.""" + +import asyncio +import hashlib +import json +from pathlib import Path +import re +import shlex +import uuid + +from harbor.agents.installed.base import with_prompt_template +from harbor.agents.model_connection import ModelConnectionSpec + +from .adapter import NanoPyCodeAgent +from .pi import distribution_digest + +_RUNNER = Path(__file__).with_name("_agent_runner.py").read_text() +_CLIENT = Path(__file__).with_name("_dsh_client.py").read_text() +_PYTHON = "/opt/harbor-dsh-venv/bin/python" + + +def distribution_files(root: Path) -> dict: + """Pin the complete offline wheelhouse, including source provenance.""" + files = {} + for path in sorted(root.rglob("*")): + if path.is_symlink(): + raise ValueError("DeepSeek distributions must not contain symbolic links") + if path.is_file(): + with path.open("rb") as stream: + digest = hashlib.file_digest(stream, "sha256").hexdigest() + files[path.relative_to(root).as_posix()] = { + "sha256": digest, "mode": path.stat().st_mode & 0o777, + } + manifest = json.loads((root / "distribution.json").read_text()) + required = {"version", "source_commit", "source_archive_sha256", "sdk_wheel", "runtime_wheel"} + if set(manifest) != required or not re.fullmatch(r"[0-9a-f]{40}", manifest["source_commit"]): + raise ValueError("DeepSeek distribution requires public source provenance") + if not re.fullmatch(r"[0-9a-f]{64}", manifest["source_archive_sha256"]): + raise ValueError("DeepSeek source archive SHA256 is required") + for key in ("sdk_wheel", "runtime_wheel"): + filename = manifest[key] + if Path(filename).name != filename or not filename.endswith(".whl") or f"wheels/{filename}" not in files: + raise ValueError("DeepSeek distribution requires SDK and runtime wheels") + return files + + +class DeepSeekHarnessAgent(NanoPyCodeAgent): + """Preserve the official sdk-minimal prompt, tools, and agent loop.""" + + SUPPORTS_ATIF = False + MODEL_CONNECTION = ModelConnectionSpec(api_key_envs=("DEEPSEEK_API_KEY",)) + CLI_FLAGS = [] + + def __init__(self, *args, distribution_path: str, distribution_sha256: str, + reasoning_effort: str = "high", **kwargs): + self._distribution = Path(distribution_path).expanduser().resolve() + self._distribution_sha256 = distribution_sha256 + self._manifest = json.loads((self._distribution / "distribution.json").read_text()) + if reasoning_effort not in {"off", "low", "high", "max"}: + raise ValueError("Unsupported DeepSeek reasoning effort") + self._reasoning_effort = reasoning_effort + super().__init__(*args, version=self._manifest["version"], **kwargs) + provider, separator, model = (self.model_name or "").partition("/") + if provider != "deepseek" or not separator or not model: + raise ValueError("DeepSeek Harness requires deepseek/model") + + @staticmethod + def name() -> str: + return "deepseek-harness" + + def get_version_command(self) -> str: + return "/opt/harbor-dsh-venv/bin/dsh --version" + + def parse_version(self, stdout: str) -> str: + return stdout.strip().splitlines()[-1] + + async def _install_agent(self, environment) -> None: + self.logs_dir.mkdir(parents=True, exist_ok=True) + (self.logs_dir / "process-supervision-required").write_text("deepseek-harness\n") + actual = distribution_digest(distribution_files(self._distribution)) + if actual != self._distribution_sha256: + raise ValueError("DeepSeek distribution changed after registration") + await environment.upload_dir(self._distribution, "/opt/harbor-dsh") + sdk = "/opt/harbor-dsh/wheels/" + self._manifest["sdk_wheel"] + await self.exec_as_root(environment, command=( + "set -euo pipefail; " + "if ! command -v uv >/dev/null 2>&1; then " + "curl -LsSf https://astral.sh/uv/0.9.11/install.sh | sh; fi; " + 'export PATH="$HOME/.local/bin:$PATH"; ' + "uv venv --python 3.13 /opt/harbor-dsh-venv; " + "uv pip install --python /opt/harbor-dsh-venv/bin/python " + "--no-index --find-links /opt/harbor-dsh/wheels " + shlex.quote(sdk) + )) + await self._upload_config_text( + environment, content=_CLIENT, remote_path="/opt/harbor-dsh-client.py", + filename="dsh-client.py", + ) + version = await self.exec_as_agent(environment, command=self.get_version_command()) + if self.parse_version(version.stdout) != self._manifest["version"]: + raise ValueError("Installed DeepSeek version differs from its distribution") + self._bootstrap_record.update( + deepseek_version=self._manifest["version"], distribution_sha256=actual, + source_commit=self._manifest["source_commit"], profile="sdk-minimal", + ) + probe = f"/tmp/harbor-dsh-probe-{uuid.uuid4().hex}" + command = shlex.join([ + _PYTHON, "/opt/harbor-dsh-client.py", "--home", probe + "-home", + "--model", self.model_name.split("/", 1)[1], + "--reasoning-effort", self._reasoning_effort, "--preflight", + ]) + initialized = await self.exec_as_agent( + environment, + command=shlex.join(["python3", "-I", "-c", _RUNNER, "run", probe + ".json", probe + "-result.json", command]), + env={"DEEPSEEK_API_KEY": "preflight-only", "DEEPSEEK_BASE_URL": "https://api.deepseek.com/anthropic"}, + ) + if not any(line.strip().startswith('{"initialized": true,') for line in initialized.stdout.splitlines()): + raise ValueError("DeepSeek SDK initialization did not complete") + self._bootstrap_record["process_supervision_probe"] = "passed" + self._bootstrap_record["sdk_initialization"] = "passed" + + @with_prompt_template + async def run(self, instruction, environment, context) -> None: + key = self.model_connection.api_key + if not key: + raise ValueError("DEEPSEEK_API_KEY is required for DeepSeek model work") + await self._upload_config_text( + environment, content=instruction, remote_path="/tmp/harbor-dsh-instruction.txt", + filename="dsh-instruction.txt", + ) + command = shlex.join([ + _PYTHON, "/opt/harbor-dsh-client.py", "--home", "/logs/agent/dsh", + "--model", self.model_name.split("/", 1)[1], + "--reasoning-effort", self._reasoning_effort, + "--instruction", "/tmp/harbor-dsh-instruction.txt", + ]) + " > /logs/agent/dsh-stdout.txt 2> /logs/agent/dsh-stderr.txt" + control = f"/tmp/harbor-dsh-{uuid.uuid4().hex}.json" + record = "/logs/agent/process-supervision.json" + try: + await self.exec_as_agent( + environment, + command=shlex.join(["python3", "-I", "-c", _RUNNER, "run", control, record, command]), + env={"DEEPSEEK_API_KEY": key, "DEEPSEEK_BASE_URL": "https://api.deepseek.com/anthropic"}, + ) + except BaseException: + cleanup = await asyncio.shield(asyncio.wait_for(environment.exec( + command=shlex.join(["python3", "-I", "-c", _RUNNER, "cancel", control, record]), user="root", + ), timeout=12)) + if cleanup.return_code != 0: + raise RuntimeError("DeepSeek subprocess cleanup did not confirm cancellation") + raise + + def populate_context_post_run(self, context) -> None: + context.metadata = context.metadata or {} + context.metadata["bootstrap"] = self._bootstrap_record + supervision_path = self.logs_dir / "process-supervision.json" + try: + supervision = json.loads(supervision_path.read_text()) + except (OSError, ValueError): + supervision = {"status": "missing_or_invalid"} + context.metadata["process_supervision"] = supervision + usage, missing, malformed = [], 0, 0 + path = self.logs_dir / "dsh-events.jsonl" + if path.exists(): + with path.open() as stream: + for line in stream: + try: + event = json.loads(line)["payload"]["event"] + if event["type"] not in {"assistant/message", "assistant/attempt", "compaction/summary"}: + continue + value = event["data"].get("usage") + if value is None: + missing += 1 + else: + usage.append(value) + except (ValueError, KeyError, TypeError): + malformed += 1 + context.metadata["deepseek_harness"] = { + "profile": "sdk-minimal", "source_commit": self._manifest["source_commit"], + "usage_records": len(usage), "missing_usage_records": missing, + "malformed_event_records": malformed, + "usage_is_partial": not usage or bool(missing or malformed) or supervision.get("status") != "completed" or supervision.get("return_code") != 0, + "cost_is_unknown": True, + } + if usage: + # The native Messages adapter counts uncached input separately. + context.n_input_tokens = sum(u["inputTokens"] + u.get("cacheReadTokens", 0) + u.get("cacheWriteTokens", 0) for u in usage) + context.n_output_tokens = sum(u["outputTokens"] for u in usage) + context.n_cache_tokens = sum(u.get("cacheReadTokens", 0) for u in usage) diff --git a/benchmarks/harbor/tests/test_deepseek.py b/benchmarks/harbor/tests/test_deepseek.py new file mode 100644 index 0000000..d510cdc --- /dev/null +++ b/benchmarks/harbor/tests/test_deepseek.py @@ -0,0 +1,184 @@ +"""DeepSeek provenance, official SDK driving, and accounting contracts.""" + +import asyncio +import json +from pathlib import Path +import sys +from types import SimpleNamespace + +import pytest +from harbor.models.agent.context import AgentContext + +from harbor_adapter import DeepSeekHarnessAgent +from harbor_adapter.__main__ import snapshot_deepseek, standard_config +from harbor_adapter.deepseek import distribution_files +from harbor_adapter.pi import distribution_digest + + +@pytest.fixture +def inputs(tmp_path): + root = tmp_path / "distribution" + (root / "wheels").mkdir(parents=True) + manifest = { + "version": "0.2.2-alpha.2", "source_commit": "a" * 40, + "source_archive_sha256": "b" * 64, + "sdk_wheel": "deepseek_harness_sdk-0.2.2a2-py3-none-any.whl", + "runtime_wheel": "deepseek_harness_runtime_bin-0.2.2a2-py3-none-linux_x86_64.whl", + } + (root / "distribution.json").write_text(json.dumps(manifest)) + for key in ("sdk_wheel", "runtime_wheel"): + (root / "wheels" / manifest[key]).write_bytes(b"wheel-fixture") + return {"distribution_path": str(root), + "distribution_sha256": distribution_digest(distribution_files(root)), + "reasoning_effort": "high"} + + +def config(inputs): + return {"tasks": [{"name": "org/task"}], "agents": [{ + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", "kwargs": inputs, + }], "verifier": {"import_path": "harbor_adapter:RetryingVerifier", "kwargs": { + "max_attempts": 3, "require_agent_quiescence": True, + }}} + + +def test_workflow_requires_public_inputs_and_quiescence(inputs): + assert standard_config(config(inputs)).agents[0].kwargs["reasoning_effort"] == "high" + with pytest.raises(ValueError, match="credentials"): + standard_config(config(inputs | {"api_key": "secret"})) + raw = config(inputs) + raw["verifier"]["kwargs"].pop("require_agent_quiescence") + with pytest.raises(ValueError, match="quiescence"): + standard_config(raw) + with pytest.raises(ValueError, match="reasoning effort"): + standard_config(config(inputs | {"reasoning_effort": "invented"})) + + +def test_snapshot_pins_wheels_and_source_and_detects_corruption(tmp_path, inputs): + agent = SimpleNamespace(kwargs=inputs.copy()) + result = snapshot_deepseek(agent, tmp_path / "cache", tmp_path / "record") + assert result["provenance"]["source_commit"] == "a" * 40 + root = Path(agent.kwargs["distribution_path"]) + assert root != Path(inputs["distribution_path"]) + next(root.glob("wheels/*.whl")).write_bytes(b"corrupted") + with pytest.raises(ValueError, match="corrupt"): + snapshot_deepseek(SimpleNamespace(kwargs=inputs.copy()), tmp_path / "cache", tmp_path) + + +def test_symlinks_and_missing_wheels_fail_registration(tmp_path, inputs): + root = Path(inputs["distribution_path"]) + (root / "escape").symlink_to(tmp_path) + with pytest.raises(ValueError, match="symbolic links"): + distribution_files(root) + (root / "escape").unlink() + next(root.glob("wheels/*.whl")).unlink() + with pytest.raises(ValueError, match="SDK and runtime"): + distribution_files(root) + + +class Environment: + default_user = None + + def __init__(self, cancel=False): + self.calls, self.uploads = [], [] + self.cancel = cancel + + async def exec(self, **kwargs): + self.calls.append(kwargs) + command = kwargs["command"] + if self.cancel and "dsh-instruction.txt" in command and " run " in command: + raise asyncio.CancelledError() + stdout = "" + if command.endswith("--version"): + stdout = "0.2.2-alpha.2\n" + elif "--preflight" in command: + stdout = '{"initialized": true, "model": "deepseek-flash"}\n' + return SimpleNamespace(return_code=0, stdout=stdout, stderr="") + + async def upload_dir(self, source, target): + self.uploads.append((source, target)) + + async def upload_file(self, source, target): + self.uploads.append((Path(source).read_text(), target)) + + +def test_install_initializes_native_sdk_without_model_prompt(tmp_path, inputs): + agent = DeepSeekHarnessAgent(logs_dir=tmp_path, model_name="deepseek/deepseek-flash", **inputs) + env = Environment() + asyncio.run(agent._install_agent(env)) + commands = "\n".join(call["command"] for call in env.calls) + assert "--no-index --find-links /opt/harbor-dsh/wheels" in commands + assert "--preflight" in commands + assert "--instruction" not in commands + assert agent._bootstrap_record["sdk_initialization"] == "passed" + assert (tmp_path / "process-supervision-required").exists() + + +def test_cancellation_waits_for_process_supervisor(tmp_path, inputs, monkeypatch): + monkeypatch.setenv("DEEPSEEK_API_KEY", "test-only") + agent = DeepSeekHarnessAgent(logs_dir=tmp_path, model_name="deepseek/deepseek-flash", **inputs) + env = Environment(cancel=True) + with pytest.raises(asyncio.CancelledError): + asyncio.run(agent.run("task", env, AgentContext())) + assert " cancel " in env.calls[-1]["command"] + assert env.calls[-1]["user"] == "root" + + +@pytest.mark.parametrize("reason,expected", [("completed", 0), ("max-tokens", 1), ("error", 1), (None, 1)]) +def test_sdk_driver_requires_completed_finish_reason(tmp_path, monkeypatch, reason, expected): + from harbor_adapter import _dsh_client + + calls = [] + + class Harness: + def __init__(self, **kwargs): + calls.append(kwargs) + + def __enter__(self): + return self + + def __exit__(self, *args): + pass + + def run(self, prompt, **kwargs): + calls.append(prompt) + return SimpleNamespace(session_id="benchmark", finish_reason=reason, final_response="done") + + monkeypatch.setitem(sys.modules, "deepseek_harness", SimpleNamespace(DeepSeekHarness=Harness)) + instruction = tmp_path / "instruction.txt" + instruction.write_text("original task\n") + monkeypatch.setattr(sys, "argv", ["driver", "--home", str(tmp_path / "home"), "--model", "deepseek-flash", "--instruction", str(instruction)]) + assert _dsh_client.main() == expected + assert calls[0]["profile"] == "sdk-minimal" + assert "max_tokens" not in calls[0] + assert calls[1] == "original task\n" + + +def test_usage_includes_compaction_and_marks_interrupted_records_partial(tmp_path, inputs): + agent = DeepSeekHarnessAgent(logs_dir=tmp_path, model_name="deepseek/deepseek-flash", **inputs) + events = [ + {"type": "assistant/message", "data": {"usage": {"inputTokens": 100, "outputTokens": 10, "cacheReadTokens": 30}}}, + {"type": "compaction/summary", "data": {"usage": {"inputTokens": 200, "outputTokens": 20}}}, + {"type": "assistant/attempt", "data": {}}, + ] + (tmp_path / "dsh-events.jsonl").write_text("".join(json.dumps({"payload": {"event": e}}) + "\n" for e in events)) + (tmp_path / "process-supervision.json").write_text('{"status":"cancelled","processes_stopped":true}') + context = AgentContext() + agent.populate_context_post_run(context) + assert context.n_input_tokens == 330 + assert context.n_output_tokens == 30 + assert context.n_cache_tokens == 30 + assert context.cost_usd is None + metadata = context.metadata["deepseek_harness"] + assert metadata["usage_records"] == 2 + assert metadata["usage_is_partial"] is True + assert metadata["missing_usage_records"] == 1 + + +def test_missing_usage_stays_unknown(tmp_path, inputs): + agent = DeepSeekHarnessAgent(logs_dir=tmp_path, model_name="deepseek/deepseek-flash", **inputs) + context = AgentContext() + agent.populate_context_post_run(context) + assert context.n_input_tokens is None + assert context.cost_usd is None + assert context.metadata["deepseek_harness"]["usage_is_partial"] is True From 5b503ab6d04af2f6a719ec0dbe5418bfad4bf128 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Fri, 30 Oct 2026 08:46:56 -0700 Subject: [PATCH 02/15] fix(benchmarks): give the deepseek version probe an explicit home --- benchmarks/harbor/src/harbor_adapter/deepseek.py | 2 +- benchmarks/harbor/tests/test_deepseek.py | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/benchmarks/harbor/src/harbor_adapter/deepseek.py b/benchmarks/harbor/src/harbor_adapter/deepseek.py index 0746d9d..8c64b57 100644 --- a/benchmarks/harbor/src/harbor_adapter/deepseek.py +++ b/benchmarks/harbor/src/harbor_adapter/deepseek.py @@ -69,7 +69,7 @@ def name() -> str: return "deepseek-harness" def get_version_command(self) -> str: - return "/opt/harbor-dsh-venv/bin/dsh --version" + return "DSH_HOME=/tmp/harbor-dsh-version /opt/harbor-dsh-venv/bin/dsh --version" def parse_version(self, stdout: str) -> str: return stdout.strip().splitlines()[-1] diff --git a/benchmarks/harbor/tests/test_deepseek.py b/benchmarks/harbor/tests/test_deepseek.py index d510cdc..4770e86 100644 --- a/benchmarks/harbor/tests/test_deepseek.py +++ b/benchmarks/harbor/tests/test_deepseek.py @@ -110,6 +110,7 @@ def test_install_initializes_native_sdk_without_model_prompt(tmp_path, inputs): assert "--no-index --find-links /opt/harbor-dsh/wheels" in commands assert "--preflight" in commands assert "--instruction" not in commands + assert "DSH_HOME=/tmp/harbor-dsh-version" in agent.get_version_command() assert agent._bootstrap_record["sdk_initialization"] == "passed" assert (tmp_path / "process-supervision-required").exists() From 82b355caafec2f2cdc5535d24f46a52d94e4f5b0 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Fri, 30 Oct 2026 08:50:07 -0700 Subject: [PATCH 03/15] test(benchmarks): register the deepseek harness dev34 comparison --- .../deepseek-harness-dev34-20261030/README.md | 111 ++ .../adaptive-rejection-sampler.json | 38 + .../apt-cache.conf | 8 + .../build-cython-ext.json | 38 + .../build-pov-ray.json | 38 + .../caffe-cifar-10.json | 38 + .../cancel-async-tasks.json | 38 + .../circuit-fibsqrt.json | 38 + .../cobol-modernization.json | 38 + .../constraints-scheduling.json | 38 + .../custom-memory-heap-crash.json | 38 + .../db-wal-recovery.json | 38 + .../deepseek-distribution.json | 58 + .../dependency-cache.compose.json | 100 ++ .../dependency-release-manifest.json | 58 + .../dna-assembly.json | 38 + .../experiment.json | 1512 +++++++++++++++++ .../fix-code-vulnerability.json | 38 + .../git-multibranch.json | 38 + .../image-digests.json | 312 ++++ .../install-pair-01.json | 42 + .../install-pair-02.json | 42 + .../install-pair-03.json | 42 + .../install-pair-04.json | 42 + .../install-pair-05.json | 42 + .../install-pair-06.json | 42 + .../install-pair-07.json | 42 + .../install-pair-08.json | 42 + .../install-pair-09.json | 42 + .../install-pair-10.json | 42 + .../install-pair-11.json | 42 + .../install-pair-12.json | 42 + .../install-pair-13.json | 42 + .../install-pair-14.json | 42 + .../install-pair-15.json | 42 + .../install-pair-16.json | 42 + .../install-pair-17.json | 42 + .../kv-store-grpc.json | 38 + .../llm-inference-batching-scheduler.json | 38 + .../log-summary-date-ranges.json | 38 + .../merge-diff-arc-agi-task.json | 38 + .../model-extraction-relu-logits.json | 38 + .../mteb-leaderboard.json | 38 + .../multi-source-data-merger.json | 38 + .../nginx-request-logging.json | 38 + .../openssl-selfsigned-cert.json | 38 + .../path-tracing.json | 38 + .../pypi-server.json | 38 + .../pytorch-model-cli.json | 38 + .../pytorch-model-recovery.json | 38 + .../qemu-alpine-ssh.json | 38 + .../query-optimize.json | 38 + .../regex-chess.json | 38 + .../regex-log.json | 38 + .../schemelike-metacircular-eval.json | 38 + .../selection.json | 150 ++ .../setup-evidence.json | 29 + .../smoke-install.json | 38 + .../torch-pipeline-parallelism.json | 38 + .../torch-tensor-parallelism.json | 38 + .../write-compressor.json | 38 + 61 files changed, 4382 insertions(+) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/adaptive-rejection-sampler.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/apt-cache.conf create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-cython-ext.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-pov-ray.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/caffe-cifar-10.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cancel-async-tasks.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/circuit-fibsqrt.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cobol-modernization.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/constraints-scheduling.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/custom-memory-heap-crash.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/db-wal-recovery.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/deepseek-distribution.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-release-manifest.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dna-assembly.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/fix-code-vulnerability.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/git-multibranch.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/image-digests.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-01.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-02.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-03.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-04.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-05.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-06.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-07.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-08.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-09.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-10.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-11.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-12.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-13.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-14.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-15.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-16.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-17.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/kv-store-grpc.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/llm-inference-batching-scheduler.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/log-summary-date-ranges.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/merge-diff-arc-agi-task.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-extraction-relu-logits.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/mteb-leaderboard.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/multi-source-data-merger.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/nginx-request-logging.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/openssl-selfsigned-cert.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/path-tracing.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pypi-server.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-cli.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-recovery.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/qemu-alpine-ssh.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/query-optimize.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-chess.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-log.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/schemelike-metacircular-eval.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/selection.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/smoke-install.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-pipeline-parallelism.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-tensor-parallelism.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/write-compressor.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md new file mode 100644 index 0000000..a0c3188 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md @@ -0,0 +1,111 @@ +# DeepSeek Harness on the Pi 34-task development pool + +Status: the 34-task installation phase is running; no model trials +have started. The official keyless SDK smoke and corrected WAL container +installation smoke passed. The first WAL setup failure is preserved in +[setup evidence](setup-evidence.json). + +This run uses the complete existing Pi development pool. `selection.json`, +original task revisions and file hashes, image identities, resources, and +native timeouts are preserved. No tasks are selected or replaced by outcome. +The other 55 tasks remain unopened. + +The upstream source is the user's clean local `deepseek-harness` checkout at +`f3803c96974ad4a94195fb97e5c637e9c2a3afe3` (`0.2.2-alpha.2`). The source archive +SHA256 is `0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4`. +Builds use an isolated copy under `jobs/deepseek-harness-build-f3803c9/` and the +upstream frozen pnpm lockfile. The original checkout is unchanged. + +The selected `sdk-minimal` profile follows the upstream `BENCHMARK.md` and +Python SDK guide. It retains the upstream system prompt, persistent bash, +image reading, working-directory tool, automatic compaction, request retries, +and generation defaults. Each task receives an isolated Harness home and +session. No personal instructions, extensions, settings, or previous sessions +are copied. The SDK is driven by the reusable repository adapter through the +standard Harbor entry point. + +The model is `deepseek/deepseek-flash`, with `high` reasoning and the official +Messages endpoint `https://api.deepseek.com/anthropic`. The official API guide +checked on October 30 still maps this mutable alias to DeepSeek-V4.1-Flash-1017. +Temperature and SDK `max_tokens` remain unset; the upstream provider default is +256,000 output tokens. Pi used OpenAI-compatible requests and model metadata +with a 384,000-token ceiling. This compares complete agent configurations; it +does not isolate the causal effect of the harness under identical generation +budgets. Neither configuration imposes Nano's 100-reply limit or finalization +reminders. + +All tasks require fresh installation-only checks before the model queue opens. +These checks install hash-pinned SDK/runtime wheels and their Python +dependencies, initialize the actual SDK profile/provider/model without a task +prompt, and exercise the Linux process supervisor. QEMU additionally runs its +reviewed verifier dependency profile. The other 33 verifier profiles are +explicitly `not_configured`; agent installation success does not certify them. + +The existing APT-index and original uv/Python release caches retain upstream +sources, bytes, size and SHA256 checks. Original task images and official tests +are unchanged. Maximum concurrency is two, with pair barriers. Setup allows +1,800 seconds; agent and verifier attempts retain task-native limits. Only +verifier timeouts may retry, at most three total attempts with the same output. +Whole-trial retries are disabled. The verifier requires confirmed agent +completion or cancellation before it starts. + +Review only operational metadata while the queue runs, and freeze every output +before reviewing new solution trajectories or hidden tests. Keep setup errors, +model execution errors, verifier errors, and scored failures separate. Preserve +every failed preparation or trial. Final results must include complete English +and Simplified Chinese reports, the Pi comparison, timeout interpretation, +reference-exposure findings, and accounting limitations. The previous Pi report +contains 26 official passes, six scored failures, and two unscored legacy +timeouts, including a pass with public reference exposure; do not treat it as +an uncontaminated 34-task reference score. + +Build preparation uses the upstream commands: + +```bash +pnpm install --frozen-lockfile --store-dir /path/to/shared-pnpm-store +pnpm exec tsx scripts/build-exe-for-python-sdk.ts \ + --targets=node24-linux-x64 --build-artifacts --compress=GZip +python scripts/build-python-release.py --package sdk --output-dir dist-python +python scripts/build-python-release.py --package runtime --platform linux-x64 \ + --runtime-exe dist-exe/deepseek-harness-sdk-runtime-linux-x64 \ + --output-dir dist-python +``` + +The [frozen experiment](experiment.json) registers 34 model jobs and 17 +two-task installation jobs. The [distribution manifest](deepseek-distribution.json) +pins the runtime, SDK, five dependency wheels, build provenance, and file modes +with SHA256 `d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6`. +The upstream eight artifact-build steps and official `sdk-minimal` keyless +smoke passed, and the upstream source remains clean. The frozen adapter source +is `5b503ab6d04af2f6a719ec0dbe5418bfad4bf128`, checked out at +`jobs/deepseek-harness-runtime-source`. + +The first container smoke failed before model work because the Python `dsh` +launcher requires an explicit `DSH_HOME` even for its version query. The adapter +now supplies an isolated version-probe home. The corrected smoke initialized +the actual SDK/profile/model and supervisor successfully. Both native records +remain preserved; neither is a scored task attempt. The complete Harbor suite +passed 280 tests before this small integration fix, and all 11 DeepSeek-focused +tests passed after it. + +Run from this repository root: + +```bash +export UV_PROJECT_ENVIRONMENT="$PWD/jobs/dsh-harbor-venv" +export UV_CACHE_DIR="$PWD/.cache/uv" +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json \ + --phase preflight +# Supply DEEPSEEK_API_KEY in the environment before the model phase. +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json \ + --phase model +``` + +Each job uses the standard `python -m harbor_adapter run --config ...` workflow +from the frozen source root. Native outputs and per-job input manifests remain +under `jobs/deepseek-harness-runtime-source/jobs/`. Orchestration and image-cache +events remain under `jobs/deepseek-harness-dev34-20261030-orchestration/`. +The runner refuses to overwrite or replay previously started model jobs. diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/adaptive-rejection-sampler.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/adaptive-rejection-sampler.json new file mode 100644 index 0000000..cfd8970 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/adaptive-rejection-sampler.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/adaptive-rejection-sampler", + "ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/apt-cache.conf b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/apt-cache.conf new file mode 100644 index 0000000..e191ddb --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/apt-cache.conf @@ -0,0 +1,8 @@ +// Route only public APT HTTP origins through the verified index cache. +Acquire::http::Proxy::deb.debian.org "http://127.0.0.1:18742/"; +Acquire::http::Proxy::security.debian.org "http://127.0.0.1:18742/"; +Acquire::http::Proxy::archive.debian.org "http://127.0.0.1:18742/"; +Acquire::http::Proxy::archive.ubuntu.com "http://127.0.0.1:18742/"; +Acquire::http::Proxy::security.ubuntu.com "http://127.0.0.1:18742/"; +Acquire::http::Proxy::ports.ubuntu.com "http://127.0.0.1:18742/"; +Acquire::http::Proxy::old-releases.ubuntu.com "http://127.0.0.1:18742/"; diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-cython-ext.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-cython-ext.json new file mode 100644 index 0000000..90d5652 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-cython-ext.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/build-cython-ext", + "ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-pov-ray.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-pov-ray.json new file mode 100644 index 0000000..a9dd9ce --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-pov-ray.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/build-pov-ray", + "ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/caffe-cifar-10.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/caffe-cifar-10.json new file mode 100644 index 0000000..03f90c7 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/caffe-cifar-10.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/caffe-cifar-10", + "ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cancel-async-tasks.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cancel-async-tasks.json new file mode 100644 index 0000000..612a221 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cancel-async-tasks.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/cancel-async-tasks", + "ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/circuit-fibsqrt.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/circuit-fibsqrt.json new file mode 100644 index 0000000..efc925c --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/circuit-fibsqrt.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/circuit-fibsqrt", + "ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cobol-modernization.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cobol-modernization.json new file mode 100644 index 0000000..aefdec7 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cobol-modernization.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/cobol-modernization", + "ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/constraints-scheduling.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/constraints-scheduling.json new file mode 100644 index 0000000..4e70572 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/constraints-scheduling.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/constraints-scheduling", + "ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/custom-memory-heap-crash.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/custom-memory-heap-crash.json new file mode 100644 index 0000000..4cca554 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/custom-memory-heap-crash.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/custom-memory-heap-crash", + "ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/db-wal-recovery.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/db-wal-recovery.json new file mode 100644 index 0000000..72a65d2 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/db-wal-recovery.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/db-wal-recovery", + "ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/deepseek-distribution.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/deepseek-distribution.json new file mode 100644 index 0000000..8b38172 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/deepseek-distribution.json @@ -0,0 +1,58 @@ +{ + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "provenance": { + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "source_archive_sha256": "0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4", + "origin": "https://github.com/deepseek-ai/deepseek-harness.git", + "node_version": "26.7.0", + "pnpm_version": "11.28.5", + "dependency_install": "pnpm install --frozen-lockfile --store-dir /mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/dsh-pnpm", + "runtime_build": "DSH_GATE_CONCURRENCY=4 pnpm exec tsx scripts/build-exe-for-python-sdk.ts --targets=node24-linux-x64 --build-artifacts --compress=GZip", + "source_git_metadata": "Original commit objects imported from the local source checkout after the official build required git rev-parse HEAD; source files remain unchanged.", + "source_clean_after_build": true, + "build_log_sha256": "73ad636ed60974c115c5d4dd0bc3819779908fb0f6cf75c95fe7e834e5838198", + "adapter_ref": "65d0fd389c61fc3a98d150be789c33da39e36f6a", + "sdk_minimal_smoke": { + "status": "passed", + "log_sha256": "df510602b41bd2c0a7aa83ed4867d8a5b675b43b2e5cc1748f7b1640b94bb9b4" + } + }, + "files": { + "build-provenance.json": { + "sha256": "36212ebe23e42e7cfab847266bac7a4f8a05d83bbce4e5422f7dcfd2f87d4785", + "mode": 420 + }, + "distribution.json": { + "sha256": "c8d7a7842b835f66eb6dd0230403f0b4af0a26dd0e05ea04810c9667d7ba3d57", + "mode": 420 + }, + "wheels/annotated_types-0.8.0-py3-none-any.whl": { + "sha256": "f072f4d804ea359e4eaf198b1af7a8b0943881a87f31bb764f8bf219bb9419e0", + "mode": 420 + }, + "wheels/deepseek_harness_runtime_bin-0.2.2a2-py3-none-manylinux_2_28_x86_64.whl": { + "sha256": "1fb53195ac1c6d5bee02a63f11488487df74d964ee7fb1603eb268d8116edddc", + "mode": 420 + }, + "wheels/deepseek_harness_sdk-0.2.2a2-py3-none-any.whl": { + "sha256": "282bfb4b373e8b4634447bddc066f6cc0f20da5d422ed5618f37d469823b742f", + "mode": 420 + }, + "wheels/pydantic-2.13.4-py3-none-any.whl": { + "sha256": "45a282cde31d808236fd7ea9d919b128653c8b38b393d1c4ab335c62924d9aba", + "mode": 420 + }, + "wheels/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl": { + "sha256": "9551187363ffc0de2a00b2e47c25aeaeb1020b69b668762966df15fc5659dd5a", + "mode": 420 + }, + "wheels/typing_extensions-4.16.0-py3-none-any.whl": { + "sha256": "481caa481374e813c1b176ada14e97f1f67a4539ce9cfeb3f350d78d6370c2e8", + "mode": 420 + }, + "wheels/typing_inspection-0.4.4-py3-none-any.whl": { + "sha256": "65b8397ba37ccbce054456aaccddfc91e6e3083c92824df348d96ca832f3f147", + "mode": 420 + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json new file mode 100644 index 0000000..395a923 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json @@ -0,0 +1,100 @@ +{ + "services": { + "main": { + "volumes": [ + { + "type": "bind", + "source": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/apt-cache.conf", + "target": "/etc/apt/apt.conf.d/99nanopy-index-cache", + "read_only": true + } + ], + "environment": { + "UV_INSTALLER_GITHUB_BASE_URL": "http://127.0.0.1:18743", + "UV_PYTHON_INSTALL_MIRROR": "http://127.0.0.1:18743/astral-sh/python-build-standalone/releases/download" + } + }, + "apt-index-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/apt_index_proxy.py", + "--bind", + "127.0.0.1", + "--port", + "18742", + "--cache", + "/cache", + "--workers", + "16" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "mem_limit": "256m", + "cpus": 0.5, + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/apt_index_proxy.py:/proxy/apt_index_proxy.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/apt-indexes:/cache" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18742), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + }, + "uv-release-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/uv_release_cache.py", + "serve", + "--cache", + "/cache", + "--manifest", + "/manifest.json" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "cpus": 0.5, + "mem_limit": "256m", + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/uv_release_cache.py:/proxy/uv_release_cache.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/uv-releases:/cache:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-release-manifest.json:/manifest.json:ro" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18743), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-release-manifest.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-release-manifest.json new file mode 100644 index 0000000..e762efb --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-release-manifest.json @@ -0,0 +1,58 @@ +{ + "registered_at": "2026-10-18T13:25:43.131474+00:00", + "metadata_source": "Original release metadata preserved by the source manifests; no versions or payloads substituted.", + "source_manifests": { + "benchmarks/harbor/configs/dev34-main-baseline-20261012/dependency-release-manifest-v5.json": "bdf3d3ef8b7743700c5294c4a511c686a3928c178a77fd4abcba90ee4557ffcc", + "python-agent-release-manifest.json": "78aecd7c397c469196f9d246a160bf1fc99ad2d51cb71320b7d26ebf839647ce" + }, + "assets": [ + { + "version": "0.9.11", + "filename": "uv-x86_64-unknown-linux-gnu.tar.gz", + "url": "https://github.com/astral-sh/uv/releases/download/0.9.11/uv-x86_64-unknown-linux-gnu.tar.gz", + "digest": "sha256:817c0722b437b4b45b9a7e0231616a09db76bab1b8d178ba7a9680c690db19f0", + "size": 21699161, + "asset_id": 318993196, + "release_metadata_sha256": "c1a1378ece65ce1aa8343e57bdbd305289a8fc6d3d9180e8bb6b0f69acd91413" + }, + { + "version": "0.9.5", + "filename": "uv-x86_64-unknown-linux-gnu.tar.gz", + "url": "https://github.com/astral-sh/uv/releases/download/0.9.5/uv-x86_64-unknown-linux-gnu.tar.gz", + "digest": "sha256:2cf10babba653310606f8b49876cfb679928669e7ddaa1fb41fb00ce73e64f66", + "size": 21370871, + "asset_id": 306970260, + "release_metadata_sha256": "1921b17d3370aafb690b29d1d121ee8719ef3d3dbc0c7cf3b948c3059a927a5d" + }, + { + "version": "20251014", + "filename": "cpython-3.13.9+20251014-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "url": "https://github.com/astral-sh/python-build-standalone/releases/download/20251014/cpython-3.13.9%2B20251014-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "digest": "sha256:c0ccda275948c79996e46993c2c5476ff5cca606dee530f1dea712179131b348", + "size": 33517986, + "asset_id": 304305340, + "release_metadata_sha256": "8e5db21cfcc2bcafcb2127f269f00cbfdd1f74842a685acd2a14e3c1ad6f8394", + "repository": "astral-sh/python-build-standalone" + }, + { + "version": "20251120", + "filename": "cpython-3.14.0+20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "url": "https://github.com/astral-sh/python-build-standalone/releases/download/20251120/cpython-3.14.0%2B20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "digest": "sha256:66f385305ae0eefd6b65e2cf942bc91d943a61e26fb7581e029f760a2b04f393", + "size": 35376131, + "asset_id": 318881850, + "release_metadata_sha256": "e3a737862d2dddf11c73bfef6c2f2579e9d4ce5bd271cca53790c998590dee58", + "repository": "astral-sh/python-build-standalone" + }, + { + "version": "20251120", + "filename": "cpython-3.13.9+20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "url": "https://github.com/astral-sh/python-build-standalone/releases/download/20251120/cpython-3.13.9%2B20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "digest": "sha256:1bc9a94ab2425fb6ae58e45f69cab61647759b57895649701390822d976fa923", + "size": 34257108, + "asset_id": 318881608, + "release_metadata_sha256": "705690ab2fd466e90b4c868635c529f3c7fd0cb597d399e350f399e3f3e35e6c", + "repository": "astral-sh/python-build-standalone" + } + ] +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dna-assembly.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dna-assembly.json new file mode 100644 index 0000000..a968f66 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dna-assembly.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/dna-assembly", + "ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json new file mode 100644 index 0000000..bba6fd2 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json @@ -0,0 +1,1512 @@ +{ + "schema_version": 1, + "registered_at": "2026-10-30T15:44:18.128749+00:00", + "experiment": "deepseek-harness-dev34-20261030", + "total_model_runs": 34, + "max_concurrent_trials": 2, + "whole_trial_retries": 0, + "minimum_free_disk_bytes": 16106127360, + "log_dir": "jobs/deepseek-harness-dev34-20261030-orchestration", + "arms": { + "deepseek": { + "agent": "deepseek-harness", + "source_root": "jobs/deepseek-harness-runtime-source", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6" + } + }, + "tasks": [ + { + "name": "terminal-bench/adaptive-rejection-sampler", + "ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/adaptive-rejection-sampler:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "scientific-computing", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "95408217b9c484ffc18a02a91b7706323b6902919c72b24866a32e27a23543e2", + "environment/Dockerfile": "e644a3343e339303230f1441b7383f076d21c69a5a7aa6430b4f8a9d27f15c8b", + "environment/protected.tar.gz.enc": "7fc9bf1347bb97da8dede04b2c58638f225070913e222d76627068aebd1b1df2", + "instruction.md": "e6d28c28acd76bef61817427c74c2a29e3e8d75f535cbd708afcbfd02c54013f", + "solution/solve.sh": "fe443a8082b5183ba1623311ba0ee958888aefa633eff93766b9ae14e9bdd84d", + "task.toml": "6ed7b37f74beb46e9d47b0f5fad53025667c9548baff4e96d467b56d429fab1a", + "tests/test.sh": "38b43560d173cc2b952c3a3e17b8a480216d84e33450515e047bcb0d806b1e0a", + "tests/test_outputs.py": "547dc6e107f034f41703722aeceb6d0236e3fb69116fc3f2fbaba11884de352f" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/build-cython-ext", + "ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/build-cython-ext:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "debugging", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "d5f5113941c3963292bae396baa9369571026f46e59437f98ddcc93475dd7729", + "environment/Dockerfile": "cb6cf421cb9b69d43bd25cd491e49bda982fbb865d47515178d017eda727194e", + "instruction.md": "7e52874e45f88505bd21540229719de6639a401477557d814489981d472f0e34", + "solution/solve.sh": "41bbdde88617070366a70e518eedca466a4ea1acd7e96899b2fb7da988e40abc", + "task.toml": "c472f0e7522c128fb7f03e00b50c8cdf2e5e6474daa1e094e7441f4b54e770b2", + "tests/test.sh": "0156c9c4085258b696ce424cf1ce96cba44d1263c5d73bdf54083fd80237f915", + "tests/test_outputs.py": "6549f1a8e04b6163ec08877e6f6b9868bf94563f3a3c72ac1fba9c512c31689e" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/build-pov-ray", + "ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb", + "native_agent_timeout_seconds": 12000, + "verifier_timeout_seconds": 12000.0, + "docker_image": "alexgshaw/build-pov-ray:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "38380f64bc631916e264f80ce4d37feedbc783310001475b367d03989d114fdf", + "environment/Dockerfile": "44f2b5fc7311d8f336daaaca99dd4216b49355e67f50b494ebacf1b0f7c14e92", + "environment/deps/illum1.pov": "5a1f7f53ee0730c4923fcd21cd4a9c34719024d8557920441551ba7d96e75dc4", + "instruction.md": "56c8ecdd0084a07f8c97986c43af5c37ed03dd5aa994da6735cf59806162d1b2", + "solution/solve.sh": "3da9dbf589656512cd7a7004dc3752d3552a6af939cfe59eff79f91695fa42c7", + "task.toml": "b90a47ec26c1191d3c523cc094ea96b170e42fbfe176ef1f814c57d05390c097", + "tests/reference_illum1.png": "bb084f430bbc9d5b623a656920670dc06cee6d9132a9d550be4a1b99e41ab9e0", + "tests/test.sh": "928d1497ad21aef0c456294e3784f1fddf8493410f23584da89f9e02ec6bd134", + "tests/test_outputs.py": "64d6ce107503714ada346e5606832ee886be9950971cf81523e73d8ba28e8d37" + }, + "prior_nanopy_work_budget_seconds": 11820, + "pi_agent_timeout_seconds": 12000 + }, + { + "name": "terminal-bench/caffe-cifar-10", + "ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf", + "native_agent_timeout_seconds": 3600, + "verifier_timeout_seconds": 1200.0, + "docker_image": "alexgshaw/caffe-cifar-10:20260403", + "resources": { + "cpus": 4, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "machine-learning", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "e918e6852c5c034422b81a6be22d7596fddcee53abe088f56727b2482da32728", + "environment/Dockerfile": "b3b8adc915a1afa035b604273af3bc1aa8b90999d2dfd72763218b34cb87abcd", + "instruction.md": "46907b0ee048f74269c29a15e0e086546b2b1220149597017da8b5ccc05a0cea", + "solution/solve.sh": "771887de1ed557269c7f3160c777519aa9538e1e9b59771b2f6ef38c8952490c", + "task.toml": "2831165ad9d9e8bbc646e6d85c0a2290640dcc4f13749547a2937c7675012a8b", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "e99b9f6284e923388fbc21488b1af99fdde46d4a809d3557b0788dbf1cb6b281" + }, + "prior_nanopy_work_budget_seconds": 3420, + "pi_agent_timeout_seconds": 3600 + }, + { + "name": "terminal-bench/cancel-async-tasks", + "ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/cancel-async-tasks:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "9adf86b1279cc44f83fe16e6d340595aa992240f1d87243d204cae32e453693b", + "environment/Dockerfile": "1ee7bdb4a7536f5f114c11a3ac6ff42790abfea901fcba6b2a8ea8685f19140e", + "instruction.md": "23d22f7039b50682941652c716f96dac314b8585c32c133614cac77dc067f47e", + "solution/solve.sh": "4938ca4ad5b7dcd36b414b42a1abb10c058ecefce7462b678f9bd05c9fae1844", + "task.toml": "14db0675e04408f2133b0067e8dbddf66a1d6e7baf6b8d38bef8ecccfc4e58a8", + "tests/test.py": "f533c0a64db068f582b4733c5b81f2e66b8fa52fcd3efa65752e6bd6d24780aa", + "tests/test.sh": "79600a6b863def31b4daa008297423010cd2d6243e689d67503e118e4dc53397", + "tests/test_outputs.py": "d6f510b51fafc81b4e9b5cc95535aa13491c8753e0e4171db42c45f4de460ee7" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/circuit-fibsqrt", + "ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3", + "native_agent_timeout_seconds": 3600, + "verifier_timeout_seconds": 3600.0, + "docker_image": "alexgshaw/circuit-fibsqrt:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "201f7c889dd6f73dd83ca078e347272b33720fa622f8c757ea6d5770453f4cbd", + "environment/Dockerfile": "780c23986778f3e5f842567868af30b1135ed095ca9449e1c90b7f0e0483d3d4", + "environment/gates.txt": "032786f94f16605767c62bb61ce4dbba7ab7d3f1046712c975f01c9a04b6ad29", + "environment/tests/sim.c": "700645ac6156cdc88ec1e210f7c7d8d9969bb42aec969c410597e8efca796173", + "environment/tests/test_outputs.py": "6117b8edb7d567b30e32b16a9244bec4c131ec01e9065d52f6f53f87e36f0438", + "instruction.md": "a7f500266a878d04ac9de39fba76220cfad216d8f6b8063bba93bebb3d4ab9ea", + "solution/solve.sh": "a4ffab4183815a0eb7646abc1de1fc67695014c060d7ff16e8bca9ec94385d0b", + "task.toml": "9f5aa1f7ebbbcec90fd109121e9bd509af3cfff6b4cec701bd91947988f40972", + "tests/sim.c": "700645ac6156cdc88ec1e210f7c7d8d9969bb42aec969c410597e8efca796173", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "6117b8edb7d567b30e32b16a9244bec4c131ec01e9065d52f6f53f87e36f0438" + }, + "prior_nanopy_work_budget_seconds": 3420, + "pi_agent_timeout_seconds": 3600 + }, + { + "name": "terminal-bench/cobol-modernization", + "ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/cobol-modernization:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "c89318cf59efcc7040e3d946022e2616161b6733931a080cdeea7b1918e7ecc4", + "environment/Dockerfile": "8b175c83fc03ec7d16d6101516bc5100230b592f480e435279f4fe051c9280b2", + "environment/data/ACCOUNTS.DAT": "46f63eb29442ebc6716bcd8a0c36f4367910f2783f6b84a307ed9a61ccd1046e", + "environment/data/BOOKS.DAT": "db04da646207cd87f8e9c5b6f0b57d95953899f3eef22de1a81423f06ef4a9fd", + "environment/data/TRANSACTIONS.DAT": "1fa4cc5420ded94d585c563e8196b85166ebbae2bb8924b20014717fb2fb9ee3", + "environment/src/INPUT.DAT": "52fc984f0cf14af1495285d7efe02b9bae13159eea59255560ee938c4f78afee", + "environment/src/program.cbl": "27cb0e56f65f52fa75f346720b37a180b8e034ebe7d7622b979f195087bd6ebb", + "instruction.md": "a87074f1141b7345394cd88053d0debb0dc8f7de95308882c6538ee9a3b359da", + "solution/solve.sh": "7d4a0504e1686c1790bfbc074d4e598f4c631f54c2bc5f3d5fd786574b4810b9", + "task.toml": "4b620211dc3d6b1b989f0210c5ecd2f606974581c8bce27bd78659d48cf0cfbd", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "009180a7092583fc8143bb3ffc3957923aa0d61d0aac3a471f76eeb4aae54a88" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/constraints-scheduling", + "ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53", + "native_agent_timeout_seconds": 1200, + "verifier_timeout_seconds": 1200.0, + "docker_image": "alexgshaw/constraints-scheduling:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "personal-assistant", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "d54a690970149e1216bd4fc066b45069c7948c13e3b1ef2a213a74342805f209", + "environment/Dockerfile": "7d9aba072a4a907229728b9739c8cf821fc2c98450babd64452682caf3759c84", + "environment/inputs/alice_calendar.ics": "5a5d45804bffd339c2479d49e76fe817b38244d14d389afe1de583443808ecbc", + "environment/inputs/bob_calendar.ics": "c97e85dbcdd756758f8a80f4db5fd5c1dcacee79298e4d2262f46de5ff680b50", + "environment/inputs/carol_calendar.ics": "32a0f8687637df0a12b8e5c56d49c4192bc3e790856943632d70f237c7c1363b", + "instruction.md": "342def78177f9bc41bea4d11af679bc385a9dda2ac8a1e425be384a5b7f937b6", + "solution/solve.sh": "802d81ec064fad3050bbade95a2f847ac24e0b7f80fa69b0bcee0347dc92fe77", + "task.toml": "7b9842de16d74e8faf8c7c0aa03b0b81145941cb12dd322d81ae23fc48607250", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "1422b345ba3aacdfc4184e962811b6379988435896c50d8cf62903e70990c5d1" + }, + "prior_nanopy_work_budget_seconds": 1020, + "pi_agent_timeout_seconds": 1200 + }, + { + "name": "terminal-bench/custom-memory-heap-crash", + "ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e", + "native_agent_timeout_seconds": 1800, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/custom-memory-heap-crash:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "debugging", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "a06ed23a2c467aa36d0d4a88e5c020b1076cc25b8d483517239658e9c596dde9", + "environment/Dockerfile": "6ee81e5a4fa2540443d7f27d577ba840728fc6aab9f301ab7919103673c78080", + "environment/patches/locale_classes.tcc.patch": "16951f4a4df6bf253886ac75f5a39fdbc328e415194493f63cb7589a364aa817", + "environment/patches/locale_init.cc.patch": "e97f133ceebd86f8881f972e0e45b0ec0a13a9bfa428d8032008707683d54292", + "environment/program/main.cpp": "b94538c76887cc2b2ab739b5ac4731cddd0faa8f4a9892a6265c062bbbeede14", + "environment/program/user.cpp": "92b203786fed1bf0b3aa1675c9cd2d82b2651170675064da8ffee73b98ef14f9", + "environment/program/user.h": "f43d155995a56c22b9cd1417daef7c3b5a976cdc662a98eef0feb438ae1eee77", + "instruction.md": "53c68ae14a3c3af9a97c40bfd9bdc30a5d98b37eaa2f69f763475745f70f108d", + "solution/solve.sh": "b0b1d10bea1cd5be14a82b35d522df616df1e67028f90d2a6bef90aaa98511a1", + "task.toml": "ed474be1983e212913aa8866ebb66216acbe36aef2339ac375d232aa46617481", + "tests/test.sh": "70a2b55f0c5757f6b123a216dfd202be4a646680bced28dddcd31fd8959a058f", + "tests/test_outputs.py": "4e1cb33f777fa7fb5e1536f9bea9eace6deccd515c67d4f87570adbe692914ba" + }, + "prior_nanopy_work_budget_seconds": 1620, + "pi_agent_timeout_seconds": 1800 + }, + { + "name": "terminal-bench/db-wal-recovery", + "ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/db-wal-recovery:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "file-operations", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "600762031a1a02ecc3b46d70f7683f3ff930b47c8a3d0e27cc31b146158c3126", + "environment/Dockerfile": "63a260531c62f4ae72060db17f295c0aa03cfc6321f8162c324b7aa6b3f81302", + "environment/main.db": "68cb288eb29e76d814b4e7ee6887b37c415675e644f88173bc191348e290d119", + "environment/main.db-wal.encrypted": "68753bdce26196a53194260d001544cc6bb8f9f38e2f2f6826d09798e31e6e26", + "instruction.md": "5a20375162921b8911c67035d1905df7fd335d1bed03044271e608b2c3c93b26", + "solution/solve.sh": "ed8530c94103bf19783a1d41bc63fe4db478cdd9629f22fee25c45a45699f5e8", + "task.toml": "1cdc2c615e259bc10265efbe71b518332070c1c002e226566399e9788f18d0aa", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "88dc5c2a3abba75bca0e0f9136223ae5bb585d5984b8d9eba51a7829c1d2ba49" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/dna-assembly", + "ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f", + "native_agent_timeout_seconds": 1800, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/dna-assembly:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "scientific-computing", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "e7855c23d9822f688eec4dd89ca0f1caf91c6a69e1c8f3c9eb544b4ab2a94821", + "environment/Dockerfile": "26e9e43a98f0b1d132bb638f44b7dc882b0d0ca32d56fa33048adeee00ab600d", + "environment/sequences.fasta": "ebdc361b5fba28eb0a2c7a8a8a444ccb12cc842c8cbc6df5a2c24254f130736f", + "instruction.md": "88998c98beabdd966ab46c04ede822bbc74796696aa1bf87c567a9131210533c", + "solution/solve.sh": "c976cb118c8ef1befd846ba05fd5cadb1b1f529bc76b9f2ea9cb3049e6e11a47", + "task.toml": "7b571544b64d9be1c91526837e5d805c2181afe0e6fd57a7f0676bba36c868d7", + "tests/test.sh": "271c8317c98539e3b49233d017fe51fb0e5fba8114e76830fe698565499b2b1d", + "tests/test_outputs.py": "5c2d0fd2e5f6303263e9750f67acda7db491ef04841832b4fc62bcfb61b08924" + }, + "prior_nanopy_work_budget_seconds": 1620, + "pi_agent_timeout_seconds": 1800 + }, + { + "name": "terminal-bench/fix-code-vulnerability", + "ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/fix-code-vulnerability:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "security", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "87f28dbad183c3a634887696d328ca68a8a0858f132bc48cb7341c22267d38f7", + "environment/Dockerfile": "0917470986224df4daf4854a71302ae81c73ba5dd6e25be1232d29b758a5b3ab", + "instruction.md": "89b1abdf0af19399f720233b5cdf6b2e8e0ff7b6e3fe3e2a3af0d4d298a8641a", + "solution/solve.sh": "7a0131b369d83ae8546ecc3f9b4e23d38240f1bd1f0941e4430fa85b5182ee03", + "task.toml": "9aa2bd14ce9408289d199d0608bc3fa65b8b5e23374c9641020de62197dbf1e2", + "tests/test.sh": "f1611da693cb4cfc5e8482698c6a1858a5c67220ae9db89b50aed66584e9b17f", + "tests/test_outputs.py": "c006a5de95b69bd78055aa26cc055734dbd50718ede25d331f767bfee4ef7f26" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/git-multibranch", + "ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/git-multibranch:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "system-administration", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "3b9473429381d9eacabab498cb4a8915aa2bca24fa857ef862e5991d3f81f113", + "environment/Dockerfile": "88d9a9f4fef1ae7fddf632799dedf5a2fb6489460a93b610c33dd8b122982821", + "environment/default.conf": "a3a4c5d0ac9cfe79b0510e4225c80fa3988c27fbac13622505886bf66449e04a", + "instruction.md": "54014a38d7526d65faeb6c11d9b7ed3d4a5cb5d973d68d3da3b68b8c0289e125", + "solution/solve.sh": "300ca96b79db8d77f14742109aaf76048aafc4644ced750b1bc5c2761e8d0175", + "task.toml": "4b9981604578a31205257cca36a28d0c89c9d8d1058541330acd11d3f8d65f82", + "tests/test.sh": "ddbde3eea6ffa8fda105cbd86597a53bd409f41e13223900aa27e77a84b6b155", + "tests/test_outputs.py": "f03c6867663b8e8fb7899df51ee4df26411fab62ed491753964b801e591d16c5" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/kv-store-grpc", + "ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/kv-store-grpc:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "51fdc2ad06f0adb1270164f247ea41799b2ee2ad15490e48ddbf5e79d3767aaa", + "environment/Dockerfile": "47c9982789092437f950736110b2bd18899c39b8fd20f3af3c95b848e564a197", + "instruction.md": "f4b3facc199d5b1c78db6e603ffa29ec1b93a77ff7d291e1a92b42e6bf582bed", + "solution/solve.sh": "6bee2938ac9731d1156b0ca02b28c3d5793bf2cbe5f995c863d397093af01cb7", + "task.toml": "6ff0f30a526bfff86a67b0b53439b9965229c0f368167dc1a5d05f34d82299a3", + "tests/test.sh": "c348467a076bf614d0a850e25b2263f85bbf7b4c7a8d344fbd5454a0611a0df8", + "tests/test_outputs.py": "115de2e964bbe689d07930527113422ef8a9e4be4eb958ad84594e3fcd2b1b0e" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/llm-inference-batching-scheduler", + "ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67", + "native_agent_timeout_seconds": 1800, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/llm-inference-batching-scheduler:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "machine-learning", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "ecc9dd635357c184742cf72fc1e452a8946e207b183fc9a7eebbcf07133b911f", + "environment/Dockerfile": "10a6193904ee0022ac098504182989893b196dfc74e626a3ca48534a8eacde48", + "environment/task_file/input_data/requests_bucket_1.jsonl": "8b3cf1e2b06074162038064dd2c114abe308acd32cec7d365ee63ee81978f7de", + "environment/task_file/input_data/requests_bucket_2.jsonl": "e15e2d4c4d55f875c42a6042610d6282f6c0657c39d3e0d0f087268252525cb5", + "environment/task_file/scripts/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "environment/task_file/scripts/baseline_packer.py": "547d230d5e6d93197803480cb81cab0ec6ac63fcfa0a24b246f549523f216c4e", + "environment/task_file/scripts/cost_model.py": "0cf417fd5ff0112defcd46970d37a290c0763c5a9092e91cb3ebb86c3f20e6ce", + "instruction.md": "0817374767533fbf51bb93345ee1bba266263c216c6cc55c096e71b4a1d37dfc", + "solution/solve.sh": "cb9efb586d76a893c3027e62029e286eb6c7583dc92dbf1840a3adc236a926d7", + "task.toml": "52aaac942f917e6a91ff658e0272195bd1cd842d41d71e302414c52603cc588b", + "tests/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "tests/cost_model_for_tests.py": "d3e810a99d40327fb8317f6512d879c52c6aeebdf5dcdc0c5e117c7df61c9ab9", + "tests/test.sh": "005f980332355e2fe143d4ff4aac11828adf070594bbe7055399c238f06953ca", + "tests/test_outputs.py": "80b06b7c4ba0db28832b767d48be74bb462d7f5df5798d9cb58c4f1160d96ffa" + }, + "prior_nanopy_work_budget_seconds": 1620, + "pi_agent_timeout_seconds": 1800 + }, + { + "name": "terminal-bench/log-summary-date-ranges", + "ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/log-summary-date-ranges:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-processing", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "026000415cb577f3d3d5132902670262f7d5d8f69841903ec06a882292f756b4", + "environment/Dockerfile": "b479f6e987b63a6ec118fa1691b8bb85c437b2a6a0fe63feab5ed7154957fdfc", + "environment/log_generator_deterministic.py": "beb6eff59979e9aa0be63ce8a0f1b0df1178f92871f04ba9f76fa70cf8beb81c", + "instruction.md": "c1d4516ad4ec2238209740eaf01291df9ab48b0028a3fc5df383c6f831453484", + "solution/solve.sh": "9f6a8ee8947bdf1c0f31b530b6194ae998d66841a3b785fa2e246ee840350ab6", + "task.toml": "9a9593673d6805ef3a301ea734e9c4dbaf80da1b37c0fe9d1f2ceea2a8c57e18", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "2fc06a6ce1944ad82f84fb437395fa75ed8c0eadd385b92e375c27e5bc9cb2b4" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/merge-diff-arc-agi-task", + "ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/merge-diff-arc-agi-task:20251031", + "resources": { + "cpus": 1, + "memory_mb": 4096, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "debugging", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "d6b48ed0890b12eb68f72c12cd02689bd6bfae2191dd2688949cde6fe77f6a94", + "environment/Dockerfile": "4a03ae6f0c60b9c2f45c33f0413b5aa75c9811870254e3ff1968fd0119173fc6", + "environment/task-deps/bundle1.bundle": "a759f9b6ffc1a2eee50778d72a1899ad7d724becab84fa30e7aac71ee9e443a5", + "environment/task-deps/bundle2.bundle": "75e3e3916a01c980a03715de67b16c729e03661d94a694b5f4223d229bd6593a", + "environment/task-deps/examples.json": "5426569e4ea078bbb74039ec712a2d4d35f1f9fa3a5f26bf2a0de4a2c8ef5db8", + "instruction.md": "2d826479873e5ac1c94c1f802763bf85c22bb04624384685ebdf6b153fe0278e", + "solution/solve.sh": "48d4becc41fda5b6063fe96947474ffe778a29d6924d7b2bb235eb61e5ce555d", + "task.toml": "6ced40e06b43b169857509765eb7f07791db558f9437cc7367d2536771c2ca5b", + "tests/test.json": "20f116a6b718bbf4874bbe203d15a5c0f85a1d88cb622e2d7529c7e4fb2d4c28", + "tests/test.sh": "c9052954cd6ce25c4a375a2fb48e765b1efa5b39f7991b93993cc793af8ab9d2", + "tests/test_outputs.py": "2754ba6c5c5f75c9db5ed2c5729d9f5d7e21cb7fdb18aa0c65abdc3e7e85c149" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/model-extraction-relu-logits", + "ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/model-extraction-relu-logits:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "mathematics", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "9465accba6c57e5de9a196eed776a16eeb0852a8afc18fc3f8979e2de71b0c7b", + "environment/Dockerfile": "598b1f80d3b7194e97cbf02c8d245de30e303909e0cbe6465125d1d3179fddb5", + "environment/forward.py": "af1a23efd8e4390d16a29777c28116b2f8623ae3f7b628fe14db9f94dac26f5b", + "instruction.md": "c9e179dd099065a34c362a865b0eb9fab5f7d96be582cf117e7519580c036ebe", + "solution/solve.sh": "4cea5094f074ff21453d4423aa1b415a104ed90caa55a4acbb4f09941a3994ea", + "task.toml": "eb64dfaf8ae88868911ae28855cdc4f767de464001e2704919015b595be6dc29", + "tests/test.sh": "3b2d8f5e0353c2bab3515d6387468a863a4f9f241ee131c73f545f3fdf9e2d52", + "tests/test_outputs.py": "e5189c7eb9942ede55e1b48a80388590b80a486b75497613f257b3373ad9b2c9" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/mteb-leaderboard", + "ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c", + "native_agent_timeout_seconds": 3600, + "verifier_timeout_seconds": 3600.0, + "docker_image": "alexgshaw/mteb-leaderboard:20260430", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-science", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "ed147a5bc31e9865b16848b2f01d2f9b1e677af57a7c632d7ea2311c0553ca76", + "environment/Dockerfile": "271fdb5b34520c0bf2cb6462365363ec52058f127cd9eea3f5da2589df5c1044", + "instruction.md": "06ea44106eca0405c25bab00f5b3cfeb611c872093f6c0f82ed057839d1d13de", + "solution/solve.sh": "4ed68831d4acf5d80d2d73eec1f21efd0896a6f51f23020f572cfc0af6da0410", + "task.toml": "86574bd1c6c4fefd2e64eee9c2caa53d87c4bcf2e51547912440a0ddbd0badb8", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "b28f8fc8b7614801305813c402be52f4be1e75de6658ecdfdf9383aa793892cf" + }, + "prior_nanopy_work_budget_seconds": 3420, + "pi_agent_timeout_seconds": 3600 + }, + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/multi-source-data-merger:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-processing", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "9bf37162a4c1b88ed93627830bb8512ebae154325d7cb069c883144a0e4c69d6", + "environment/Dockerfile": "c95a40d3d35b473dc15ae1e8784ba7fb1725805c0583fc574dd2764325d75e4f", + "environment/data/source_a/users.json": "75e51e2d3ab0bdba2269d7cf3757576f84f3ee2c13407d2db94289a5958a1778", + "environment/data/source_b/users.csv": "8e87d16dc47d166edd8e56165955bad265b630d53d02762367c816affb5fa760", + "environment/data/source_c/users.parquet": "23d1ea51c4170bdba53febb885a0f495e28de525a52667347b2932136b9a72fe", + "instruction.md": "d1d28bb6e37f7a255a1b7f78aae8591f35c0494f33e4913392424ad954aab944", + "solution/solve.sh": "1b3d0d50e31930d042495d48f4c8e95f1a9c1a979c803b7f7cec651322537e1e", + "task.toml": "ee3964c0ee040f0a0cb121cefefbc15443d127684acb003fa5b61dd42f31c694", + "tests/test.sh": "d5a00e64413c0e1ae452f38a1aa1a47c3ed7b4416bbfccae1a4c053025a336d1", + "tests/test_outputs.py": "a070888930cf829aeca039771260b1386bda990ea805f313468d381ba3608ca5" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/nginx-request-logging", + "ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/nginx-request-logging:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "system-administration", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "d79997586d1c14ed8dbfea050e3580b25500e04c8345acfd3f294152433c795a", + "environment/Dockerfile": "43003466d89ab83484a3fa618e35b58cdb74a002b1e8716ef30f12b3ac8302d8", + "instruction.md": "26a7ac98aced6107f147206790ada77bbf8ed3c25fb360cf323d5fc6889edf99", + "solution/solve.sh": "2454bbbf7c201a9a778165f286373ba2cb991a0eb1d026a8005d57ae71250dd9", + "task.toml": "b7a78d00be13656f75069197100b7b8c4fd268decf6f78929c853b3222a2729d", + "tests/test.sh": "27761195f53cbfe5941fe39b573631f6c926869e64b7c43220377e1e116528e7", + "tests/test_outputs.py": "045cc716c14efde3b0dcff5fc7c85ec5d18bfc6ce66f8b40a418fa2a3a4acda0" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/openssl-selfsigned-cert", + "ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/openssl-selfsigned-cert:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "security", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "5dfa88651890cda06fa2a25babd3b813818767635e88bbabecc8305d5bb3942f", + "environment/Dockerfile": "f590b523038be43afcaf5291d7b9e8ed1333c48d9994e77b8dee2f2c352f0f40", + "instruction.md": "38054a3371f7c2c853478a2f9af0bff858437d5b28a9a07483238f06e566921f", + "solution/solve.sh": "58cf3c844e907745cd81fb4bd75efb3d080e2f0e82b589c56c954ae6e1756510", + "task.toml": "75c8913f0fa15890786c11999cde920b8bfc10d4ce8a14edf1e76985a2ac5800", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "90970bb3c372e8e7fb6bb9ac5f649ba73eb38e1a5ea2e922a9057a72b6b4192d" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/path-tracing", + "ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c", + "native_agent_timeout_seconds": 1800, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/path-tracing:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "3e4fc952b3f9a860c7ee7837b3f3c686f1283ee7821b014265dba7c94d7fc849", + "environment/Dockerfile": "7aca02e4e20b04ed2269d627a29dd4e7337a02764a9652b641452bf09836a3d5", + "environment/orig.c": "665c1bde0907740890ac114d6b36af1f3fd00449fa287583817f778bb5da200c", + "instruction.md": "9f36e02d636f5dc0946619bc518f2e51c24fa75251da9d1f0ca9eca612cb9572", + "solution/solve.sh": "d107abde101eecee6838da80cc52a3d480c93956fff3a9343a299ea01b3fe9e6", + "task.toml": "cf47013f69bf1b43242da984f231d4daf364fdf44d3845f2eea2ca2cb9308c46", + "tests/test.sh": "a719dbb1ba7ae76df095e9d815941dfa4c55edaf11f173b9c3c3097afa838548", + "tests/test_outputs.py": "572e68b2843191d29592faf42a405861b7958d8e894868b74c5a8886b53eb879" + }, + "prior_nanopy_work_budget_seconds": 1620, + "pi_agent_timeout_seconds": 1800 + }, + { + "name": "terminal-bench/pypi-server", + "ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pypi-server:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "906a096a2cc7b9afc738ff5ad5fbd605be65c103f5069c6673915223b9b187e3", + "environment/Dockerfile": "17a4e67580e4f2c6a151d04b9d74eb020666e7eec8e5259c4a5411f9e93858d0", + "instruction.md": "1c537e4a311a8e71155e121f57bfc03f742fbfb99d05314cd03e433996a28a31", + "solution/solve.sh": "1143c0eade4932ec3aeaa5013fb5e23bbcccca15dead294aa03b6af88750775a", + "task.toml": "07c39d7a733c89adadf74bb63a3e906c3694feb6c490d5fc85ac187c003fd447", + "tests/test.sh": "d2b1220e42bcfdcce87907ac5bbd69ac210a3c94926545ca2439636786e7784e", + "tests/test_outputs.py": "82c6351722e9aa984ff1d701bf57df3150ce51466df2b70afa1aa62d2650be4c" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pytorch-model-cli:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "model-training", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "35a093fa5f1592737784ba718c2e3e9ae17ff8b5e53e003fae560ddce091039c", + "environment/Dockerfile": "8fd5c0f21a6914cd440d10c84c9db06fec606f20dc3e0363b17a3b8c220dc4d9", + "environment/task-deps/cJSON.c": "0e7995261640c96db16ed2c91a5d981efa917cf5e72d0f97ec45a1179ae3ad73", + "environment/task-deps/cJSON.h": "0909160541bcc510e0b2d8c7a9cc051fa1459a0f13be4d8e56346a15491f2f2f", + "environment/task-deps/image.png": "f4d5dfefe463939996541404340e7099d5957b56e1df3760d13b5d2de754db53", + "environment/task-deps/lodepng.cpp": "67873c5f6272f779a143a9c81070e8783327c5552d0b9de2aaee2cb779308917", + "environment/task-deps/lodepng.h": "b735b7e0357319973e9488d520d7f270d734a6e9e1bf924fb1d5b8d26d7e97ce", + "environment/task-deps/model.pth": "a32103eef09936e586b1a7dc7e64d279500a38ff30fdd6df2167e28eb954eee6", + "environment/task-deps/model.py": "e4d3a5c728449656d0efde7871ee227619348352769795ed1be1a8b7f835fddc", + "instruction.md": "ce36ee1f6c589f81ee6a6be4967e25c76a2ed697006928023f42dfb9f28a6750", + "solution/solve.sh": "3b34e561c9ef0a53c9948655847846eb4811aaac67b990480aa6aac1b04bfe66", + "task.toml": "fc1c018efbd5d0537ac0741d53d1ebad3886faa51d14cbde8ec953856411bc0d", + "tests/test.sh": "43fb67e537c7adc2df11c2367f47c876cb5d16a7c1ae16d0339c3a3364bb2b06", + "tests/test_outputs.py": "370702c57a4e0c85f70ee37a5c357ae3a5df8480a275bb6263b0376f2f6c093b" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pytorch-model-recovery:20260430", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "model-training", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "e080308a05703b40edd16f53aea41a0e6b9874750ecee35bb432029d0d44bf53", + "environment/Dockerfile": "8a6b7edbb9e7c241e447f15a71e13e5d043b25169be2fd9802b84f5be3d56eb6", + "environment/dataset.pt": "6db65e4ec1c0760209c2dd997ef582e4a1519ab38e0312339999c460a696960d", + "environment/weights.pt": "71471e71c5a311a897362dd5b77413e7ca2ca59abd00a110bdc2cd3f8584d080", + "instruction.md": "2bef07dc28be3ab2c35ae389584034c2e4dde92c2c31b3eb6bd829628c8c3f4a", + "solution/solve.sh": "175a7802777b43e0fa4f629d71f7c68fb40706813c342aaddc72108d89bd5b2c", + "task.toml": "851a51204f53482a31f47cdf8833128cf421113b0331011286940fc857c37dfd", + "tests/test.sh": "6005f3a75ccfe00d7bed1005264b4a84a1f8e30cd780f8a97b3a804be7cc7a8e", + "tests/test_outputs.py": "5d4d57c43e6913bf3d0081da764ad36d1b7275e3bb114a05abda529afe563905", + "tests/weights_gtruth.pt": "71471e71c5a311a897362dd5b77413e7ca2ca59abd00a110bdc2cd3f8584d080" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/qemu-alpine-ssh", + "ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/qemu-alpine-ssh:20251031", + "resources": { + "cpus": 1, + "memory_mb": 4096, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "system-administration", + "verifier_dependency_profile": "qemu-alpine-ssh-v1", + "task_file_sha256": { + "README.md": "72f1c83fa1e3e6381ef85dd82fd109ee18cdeac090311cf3212a0a8cff0a32a5", + "environment/Dockerfile": "015ce0601ff54758f66b33f4952e0d3a577d84c09aed8f3ec5945a570e8d168a", + "instruction.md": "8348c60490cb90bc690cba259dfe98996da4c3294c6d42bba1119b00c64de4de", + "solution/solve.sh": "1cd31186c427b91bb00499d22aa4f0214fc1ca93b40b6e4cfa969611b848db2a", + "task.toml": "10a43ec6ee732eeba852c377f12e23178f7fa5a4868143f3699a15565ffb64c7", + "tests/test.sh": "3b244479776c4ee3639631112799f2cde22cbb80ac10511715e3a8f32909c289", + "tests/test_outputs.py": "83ccb4bdcd259b342212bebefe395f289858b4dc55b2ee7ae0dab238ad2e9718" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/query-optimize:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-science", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "a82c53e333a0ac0b0b410b5e26cb8e7305dece31db84d340248ffb13d5a45160", + "environment/Dockerfile": "064818771f200e0517d22068f87229a7a95d0a82b0778c84ca7203bf1926a0b8", + "environment/task-deps/my-sql-query.sql": "43382792f0f438625d643e34abe144febd5669518f4ab1ab550a15c6d9314a21", + "instruction.md": "473e12d320d2732f679fb631c2a4dab4a01deea6775686718ae0c417b8a39447", + "solution/solve.sh": "0d95f3cedb287ff9d47c796cbaf812c300d1222c10fa80c49e8acb0e81eb6978", + "task.toml": "6e88f2a86df699594bcb1bf81df4ed694505f79c9420f671335b02524bbe3868", + "tests/golden.sql": "4da1885d56b3ffa7b7b22b422996d018795611f6963499d551582881b74b84fd", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "2068ae80c57af8d52a7ae510f0666732c057806e556b480970f36f14bae683f6" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/regex-chess", + "ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78", + "native_agent_timeout_seconds": 3600, + "verifier_timeout_seconds": 3600.0, + "docker_image": "alexgshaw/regex-chess:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "9296cbee592fc996c58d228b51fee303647e5ecfae82d5666d42a40cb30ec355", + "environment/Dockerfile": "2616cd775252303839dfda0ee2f047eb862952b053b0103bd6b2460ef90c11b7", + "environment/check.py": "f9c416ae9ea01dcdd5f6a43170c8434b8c0a0b389a4a30876a2048e277d191c7", + "instruction.md": "a6ed3fcc54fc92d71a84af921942422d299ca89b410dcd92cd05a53d0d06436c", + "solution/solve.sh": "b728bf0ea392a3d244afbd0818d36933c8d3e190e47743c1382cd518d518a2e9", + "task.toml": "07fe52636fcc4bc9fff25d5630af145b97a5331219d719cc8decb4ca67ea21d4", + "tests/test.sh": "8f7f00a1c4dde7fcca6c918c768f2aba432b8c4007c15989716c68091950da87", + "tests/test_outputs.py": "4983c44d21c56dd8191381eff0304afc8b87bb714527ea9036b3319c015a30b4" + }, + "prior_nanopy_work_budget_seconds": 3420, + "pi_agent_timeout_seconds": 3600 + }, + { + "name": "terminal-bench/regex-log", + "ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/regex-log:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-processing", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "29ffd98c55dbb6f4844f1b9d8c1769b2155c117ee475e0df6bbc5ef330d239b3", + "environment/Dockerfile": "5f264ff0bf62a49aee3c6398447de5108df141e7a2ad35c6de8ef39fa62049c6", + "instruction.md": "4f7ac05e70cf9220ea0f1e5a052c5f908cd0fa884e847d80b0bd51bae2e96f9c", + "solution/solve.sh": "7e670d4f2b4bccb1e4db38f2a173e085ceda028c38167912b466b0a84fcc0999", + "task.toml": "6bc98525852755caf1fe34c174c2f10f4328adb4ce6c1968dcafe1c877e4e160", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "345c3bd09ab6f6fe8c8361a58c0a47bf0a13b3fcb38a5ac7824e44ff855e8f72" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/schemelike-metacircular-eval", + "ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6", + "native_agent_timeout_seconds": 2400, + "verifier_timeout_seconds": 2400.0, + "docker_image": "alexgshaw/schemelike-metacircular-eval:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "095d3ec6dcc0fe0c7a7e92355456dcea6992f91d1f8d397669b8cfe18475f506", + "environment/Dockerfile": "9b3df76222e16aad360fbd34cb0f71a0be8986052c065b10b92c0cdf417b25c7", + "environment/tests/interp.py": "efb8b5cdcf78b73afcf39d59600a1f355ca89b7bac9854bb01f110cc77bea2c1", + "environment/tests/shadow_test/01-factorial.scm": "540e81fac476f04dc510ba107721d606ac980760088b2136889e7d4bf884e5c5", + "environment/tests/shadow_test/02-fibonacci.scm": "628957171e0c4a3ee373635923fcd29e75bdf4bf13a402ac26fe94a5fc1443ac", + "environment/tests/shadow_test/03-list-operations.scm": "fe8f17f59fe4045582fce363d52df95ba7c4de7345542c8bd4993a9f2f396235", + "environment/tests/shadow_test/04-higher-order.scm": "9c2eba8c91bb7b224de56d789207d2eaad98bbe1f8c360d6eca4f014f42f9230", + "environment/tests/shadow_test/05-simple-io.scm": "1eba30db30824fcf6048ea8e49b84e85c866cefce30862fce49b300e4416843c", + "environment/tests/shadow_test/06-interactive-io.scm": "4fbc0e2d6e42c5a9495e549ccb13c02fcaeddf8fdaf8d2cf5de833116f12610a", + "environment/tests/shadow_test/08-progn-sequencing.scm": "3bf8236dc5356d29ab8dbfe1b0f4c519354fbb23c8f3eaf946779c28b80feea3", + "environment/tests/shadow_test/09-mutual-recursion.scm": "3bfa1f77e602c7039f8c8c70315e6d4fff6eed737dab0f0398e5bdf690137c2c", + "environment/tests/shadow_test/10-advanced-features.scm": "b69ef497088fcf0aec388789a17561b783b8cdd5915719915a51c5ba189c31ff", + "environment/tests/shadow_test/accumulator_patterns.scm": "8b9192796a6f9c3155cc5e44d0a0ba6a46d20a0d94e265b69b23ec8af43e4a07", + "environment/tests/shadow_test/binary_tree.scm": "5a3dafc8844969cebb5809f253f4e54de7df480f378da8197f870badef03ee47", + "environment/tests/shadow_test/church_numerals.scm": "31d075785f2f605bac3bd9ac2d99fcf7eaaed6686eb9afeab1d40bf5f35bbf2c", + "environment/tests/shadow_test/closures.scm": "c7d025a376987f5ee22f529847b66eaf53f0af7033acbb4055cedb734168d612", + "environment/tests/shadow_test/continuation_passing.scm": "79e7445c42227846f1adfaf08c17e278aac7a34df77482900b068be4fd248080", + "environment/tests/shadow_test/currying.scm": "690a8cccbb1d249c8f6fcabc0193cecc5d0e4c1af39a0bb67c7e95fcff8e393e", + "environment/tests/shadow_test/filter_operations.scm": "485a0563b81d002e368c5e4b2b7cc98a5306163701053630f416e3b783a2a238", + "environment/tests/shadow_test/fold_operations.scm": "461b33db6b3f5580122d4c6dff0fd6d11c7f044865ffd684ade67d9b296a5a24", + "environment/tests/shadow_test/function_composition.scm": "65ceee00188b12c3549134449e2f938d010c2c97dba090f5b53244abb3579b1f", + "environment/tests/shadow_test/lazy_evaluation.scm": "edd95748de14aadf916c65209b5bcb8759d3e33b7ec935289a5128a656d33bec", + "environment/tests/shadow_test/list_operations.scm": "660963851ecb49639f66557bd88db1e73db323186742145f33cb2e14d3af1103", + "environment/tests/shadow_test/map_operations.scm": "974ac42d892f4b62b9b21ef7844581624dad811b38d35944f1005cc47f249856", + "environment/tests/shadow_test/memoization.scm": "25482c1912fab1235651b1b7acd727f2ea6a55dc5c1919f841b44cfc32107b05", + "environment/tests/shadow_test/mutual_recursion.scm": "2fbc9562e863cb2b325f86419fb2a7588265aaf5576295550722ee4274663712", + "environment/tests/shadow_test/nested_defines.scm": "23ad3b2f936949501466da784511ea2e6215e2b198ca371bef86f6af877ca923", + "environment/tests/shadow_test/oeis_sequences.scm": "1b586df6af2429a54808f7587f60021ba2b2c9bcc2ef2f9cbe44339699853c62", + "environment/tests/shadow_test/oeis_sequences2.scm": "300da0b0dc20b883e01142adea9eb6da1a583a1b969ac0be47aa32c1ab66340e", + "environment/tests/shadow_test/oeis_sequences3.scm": "dea198e874bf518b93c8406a843b64c1586cf4ce780fec8c444f8343e9322c75", + "environment/tests/shadow_test/recursive_structures.scm": "64f1a430bf04a627f5e27bf004805544b60406f8a123d186516774dfcf3846cd", + "environment/tests/shadow_test/test_read.scm": "fda448c22685e2f9af37ff7b986ee5e906d485a2db16f62eec113149f866870f", + "environment/tests/shadow_test/variadic_functions.scm": "ac940ac0273e87ee06280fee192194ab5cdc14c88b61da9cfba92105e66b8f2f", + "environment/tests/shadow_test/y_combinator.scm": "38d25b02d54342ab5610ad03fcef68c5f1d4eef8902295645f6acceb853c9860", + "environment/tests/test/01-factorial.scm": "2623000a742c63d50ae1851854826247caee129c7d1613be6c7131d21f26a6ad", + "environment/tests/test/02-fibonacci.scm": "db769cef446d5da9e9b1abccdf9ca644e239938095a98c21551aee25255a57d4", + "environment/tests/test/03-list-operations.scm": "1a14b4100dedef30010fed62ea9fd295add1b0facb2e8e6e27c8d8eb3d90111b", + "environment/tests/test/04-higher-order.scm": "59cd17bbffad641a37729b17c5bf9178b8bae7289fe89cc7135eb0bfec2b4fd6", + "environment/tests/test/05-simple-io.scm": "3deeb3275df1b771449ebaebfcb9a934f375415e138c5e9f3685516c69a1f7a0", + "environment/tests/test/06-interactive-io.scm": "aa3cae79b79dca90ac9d61a1babf64d6a5bb6559798c7727c11be02d8ddfd216", + "environment/tests/test/08-progn-sequencing.scm": "4c0a4d980cf4223d1d17715d58adbe27dc120eced257a0d9f42de952c094bc5b", + "environment/tests/test/09-mutual-recursion.scm": "939e18f4bd25b5ab0e2a4a767bf7eb6b0d8fb3530d28bf236f06cf27f2d90ee8", + "environment/tests/test/10-advanced-features.scm": "c4cc5fb412df363875768c6a1e645255f6bc8f9ddf988d1ec2f4a266cedb8ec7", + "environment/tests/test/README.md": "1f2174c7f12721f9f1898cda190c2886fb3ef9d35fd085f315da5814657104ae", + "environment/tests/test/accumulator_patterns.scm": "50d246e00577d1f3ab3f0462c2b9618ea63680c33fcca12369cd1707b010cb5f", + "environment/tests/test/binary_tree.scm": "c95f59ccac1ea4a9c846afddc993aa40ac081c0f9e6731544530109b269b67ae", + "environment/tests/test/calculator.scm": "a09213ec8a2ec52486b9346c420ddb7c537576ef71b1cc5c6ec61d13d9139f61", + "environment/tests/test/church_numerals.scm": "f37d858f32a9c5ba8868d094018c92516acadf0ea4e55b7c36f14320cb0c69a6", + "environment/tests/test/closures.scm": "3ecfc35289479288876e20bf7f55e5b8eda84acd35937e89e3d062308a99aeff", + "environment/tests/test/continuation_passing.scm": "01049581e582f7d35ef40ef68e1827d012da619beea57c1c74675a5f245318d7", + "environment/tests/test/currying.scm": "2844d54eee9a94af05a11b4aae06d6e7823d193f17235c8f90bf98926cbca75d", + "environment/tests/test/filter_operations.scm": "f6bc75f6da9ece96c0616e3570e4df59d0f2eb4655375d3a46b0349c86cc5dd8", + "environment/tests/test/fold_operations.scm": "05aab492ad914baecd19cf454ed587e7a031e94b8990500eb1c869c15da6fc0b", + "environment/tests/test/function_composition.scm": "9b56dc78823e7d35adb213744c0f8e95684074d6a12bb98c8c05755788be5b06", + "environment/tests/test/lazy_evaluation.scm": "ea5912a53a35b65bf2bee9483dca90814b8baf52ae3ae50c2d4589f794dca54d", + "environment/tests/test/list_operations.scm": "d7e70621afbbb787cf646e2089bccfbd3681fbb2e70325f3e50f56ccd30dbd9d", + "environment/tests/test/map_operations.scm": "0cebddb3d4a50e4b6f6bfc0e8c12762d3f42609030aef077d8ab4bedda4718d9", + "environment/tests/test/memoization.scm": "cf8202df22407dbba72ca7f660d8ab7cd4ca962b215c36cc491d5c1a474c160b", + "environment/tests/test/mutual_recursion.scm": "a3dc23a96ec1cf05b0be4a4f0b6d3da9e2bad1aa3f7a83c7566f71f980a67fe2", + "environment/tests/test/nested_defines.scm": "21faa0ccaf744cb8c9d283bbfd0c36e195c803c85f511e70a99267c0a85b6216", + "environment/tests/test/oeis_sequences.scm": "d1f77161b367176b3dfab55a34466e511c602c7ae359e05f5a715627b93dca6a", + "environment/tests/test/oeis_sequences2.scm": "5679ef0f99743114d8c06edcdcde1fea9474b71aeb57c2069f1ee33df1970bce", + "environment/tests/test/oeis_sequences3.scm": "5dc8b44158b2648a6e9766f6bcdf21553a07bf487d79e64ebb418511ef3b8186", + "environment/tests/test/recursive_structures.scm": "67838ff5bac26ea359139720adec1d5700176147b0062065c7def9715e8a2aeb", + "environment/tests/test/test_read.scm": "5b9ee381c3cf45b2fc8c2e7b490857fb67f1ac78653b586d7f2848daf6853809", + "environment/tests/test/variadic_functions.scm": "c9b847c50840c6977904e5033a07f6108cc3a54012b02d916e79025fb43e0f6e", + "environment/tests/test/y_combinator.scm": "971144dcb2b04aaa3b34d9279459bd7d2e8256c585d17809465848c2589d36d8", + "environment/tests/test_outputs.py": "fbb84e3c0c2f72ed7de90f54380f860052c16155e7d0c9076cd57c966e9be7ed", + "instruction.md": "3af5f9567b62f93683db7518ec6fcb609f83796258aa02166b5a7091c5b69031", + "solution/solve.sh": "e2c5b0ac2bef4801a1891e13c108813381877426c1a7f01bab86b4fc1f712510", + "task.toml": "8dae8eb226183b0ca92331c1106152a7abc93b03b4e5a95291b86a4ddc9b44e0", + "tests/interp.py": "efb8b5cdcf78b73afcf39d59600a1f355ca89b7bac9854bb01f110cc77bea2c1", + "tests/shadow_test/01-factorial.scm": "540e81fac476f04dc510ba107721d606ac980760088b2136889e7d4bf884e5c5", + "tests/shadow_test/02-fibonacci.scm": "628957171e0c4a3ee373635923fcd29e75bdf4bf13a402ac26fe94a5fc1443ac", + "tests/shadow_test/03-list-operations.scm": "fe8f17f59fe4045582fce363d52df95ba7c4de7345542c8bd4993a9f2f396235", + "tests/shadow_test/04-higher-order.scm": "9c2eba8c91bb7b224de56d789207d2eaad98bbe1f8c360d6eca4f014f42f9230", + "tests/shadow_test/05-simple-io.scm": "1eba30db30824fcf6048ea8e49b84e85c866cefce30862fce49b300e4416843c", + "tests/shadow_test/06-interactive-io.scm": "4fbc0e2d6e42c5a9495e549ccb13c02fcaeddf8fdaf8d2cf5de833116f12610a", + "tests/shadow_test/08-progn-sequencing.scm": "3bf8236dc5356d29ab8dbfe1b0f4c519354fbb23c8f3eaf946779c28b80feea3", + "tests/shadow_test/09-mutual-recursion.scm": "3bfa1f77e602c7039f8c8c70315e6d4fff6eed737dab0f0398e5bdf690137c2c", + "tests/shadow_test/10-advanced-features.scm": "b69ef497088fcf0aec388789a17561b783b8cdd5915719915a51c5ba189c31ff", + "tests/shadow_test/accumulator_patterns.scm": "8b9192796a6f9c3155cc5e44d0a0ba6a46d20a0d94e265b69b23ec8af43e4a07", + "tests/shadow_test/binary_tree.scm": "5a3dafc8844969cebb5809f253f4e54de7df480f378da8197f870badef03ee47", + "tests/shadow_test/church_numerals.scm": "31d075785f2f605bac3bd9ac2d99fcf7eaaed6686eb9afeab1d40bf5f35bbf2c", + "tests/shadow_test/closures.scm": "c7d025a376987f5ee22f529847b66eaf53f0af7033acbb4055cedb734168d612", + "tests/shadow_test/continuation_passing.scm": "79e7445c42227846f1adfaf08c17e278aac7a34df77482900b068be4fd248080", + "tests/shadow_test/currying.scm": "690a8cccbb1d249c8f6fcabc0193cecc5d0e4c1af39a0bb67c7e95fcff8e393e", + "tests/shadow_test/filter_operations.scm": "485a0563b81d002e368c5e4b2b7cc98a5306163701053630f416e3b783a2a238", + "tests/shadow_test/fold_operations.scm": "461b33db6b3f5580122d4c6dff0fd6d11c7f044865ffd684ade67d9b296a5a24", + "tests/shadow_test/function_composition.scm": "65ceee00188b12c3549134449e2f938d010c2c97dba090f5b53244abb3579b1f", + "tests/shadow_test/lazy_evaluation.scm": "edd95748de14aadf916c65209b5bcb8759d3e33b7ec935289a5128a656d33bec", + "tests/shadow_test/list_operations.scm": "660963851ecb49639f66557bd88db1e73db323186742145f33cb2e14d3af1103", + "tests/shadow_test/map_operations.scm": "974ac42d892f4b62b9b21ef7844581624dad811b38d35944f1005cc47f249856", + "tests/shadow_test/memoization.scm": "25482c1912fab1235651b1b7acd727f2ea6a55dc5c1919f841b44cfc32107b05", + "tests/shadow_test/mutual_recursion.scm": "2fbc9562e863cb2b325f86419fb2a7588265aaf5576295550722ee4274663712", + "tests/shadow_test/nested_defines.scm": "23ad3b2f936949501466da784511ea2e6215e2b198ca371bef86f6af877ca923", + "tests/shadow_test/oeis_sequences.scm": "1b586df6af2429a54808f7587f60021ba2b2c9bcc2ef2f9cbe44339699853c62", + "tests/shadow_test/oeis_sequences2.scm": "300da0b0dc20b883e01142adea9eb6da1a583a1b969ac0be47aa32c1ab66340e", + "tests/shadow_test/oeis_sequences3.scm": "dea198e874bf518b93c8406a843b64c1586cf4ce780fec8c444f8343e9322c75", + "tests/shadow_test/recursive_structures.scm": "64f1a430bf04a627f5e27bf004805544b60406f8a123d186516774dfcf3846cd", + "tests/shadow_test/test_read.scm": "fda448c22685e2f9af37ff7b986ee5e906d485a2db16f62eec113149f866870f", + "tests/shadow_test/variadic_functions.scm": "ac940ac0273e87ee06280fee192194ab5cdc14c88b61da9cfba92105e66b8f2f", + "tests/shadow_test/y_combinator.scm": "38d25b02d54342ab5610ad03fcef68c5f1d4eef8902295645f6acceb853c9860", + "tests/test/01-factorial.scm": "2623000a742c63d50ae1851854826247caee129c7d1613be6c7131d21f26a6ad", + "tests/test/02-fibonacci.scm": "db769cef446d5da9e9b1abccdf9ca644e239938095a98c21551aee25255a57d4", + "tests/test/03-list-operations.scm": "1a14b4100dedef30010fed62ea9fd295add1b0facb2e8e6e27c8d8eb3d90111b", + "tests/test/04-higher-order.scm": "59cd17bbffad641a37729b17c5bf9178b8bae7289fe89cc7135eb0bfec2b4fd6", + "tests/test/05-simple-io.scm": "3deeb3275df1b771449ebaebfcb9a934f375415e138c5e9f3685516c69a1f7a0", + "tests/test/06-interactive-io.scm": "aa3cae79b79dca90ac9d61a1babf64d6a5bb6559798c7727c11be02d8ddfd216", + "tests/test/08-progn-sequencing.scm": "4c0a4d980cf4223d1d17715d58adbe27dc120eced257a0d9f42de952c094bc5b", + "tests/test/09-mutual-recursion.scm": "939e18f4bd25b5ab0e2a4a767bf7eb6b0d8fb3530d28bf236f06cf27f2d90ee8", + "tests/test/10-advanced-features.scm": "c4cc5fb412df363875768c6a1e645255f6bc8f9ddf988d1ec2f4a266cedb8ec7", + "tests/test/README.md": "1f2174c7f12721f9f1898cda190c2886fb3ef9d35fd085f315da5814657104ae", + "tests/test/accumulator_patterns.scm": "50d246e00577d1f3ab3f0462c2b9618ea63680c33fcca12369cd1707b010cb5f", + "tests/test/binary_tree.scm": "c95f59ccac1ea4a9c846afddc993aa40ac081c0f9e6731544530109b269b67ae", + "tests/test/calculator.scm": "a09213ec8a2ec52486b9346c420ddb7c537576ef71b1cc5c6ec61d13d9139f61", + "tests/test/church_numerals.scm": "f37d858f32a9c5ba8868d094018c92516acadf0ea4e55b7c36f14320cb0c69a6", + "tests/test/closures.scm": "3ecfc35289479288876e20bf7f55e5b8eda84acd35937e89e3d062308a99aeff", + "tests/test/continuation_passing.scm": "01049581e582f7d35ef40ef68e1827d012da619beea57c1c74675a5f245318d7", + "tests/test/currying.scm": "2844d54eee9a94af05a11b4aae06d6e7823d193f17235c8f90bf98926cbca75d", + "tests/test/filter_operations.scm": "f6bc75f6da9ece96c0616e3570e4df59d0f2eb4655375d3a46b0349c86cc5dd8", + "tests/test/fold_operations.scm": "05aab492ad914baecd19cf454ed587e7a031e94b8990500eb1c869c15da6fc0b", + "tests/test/function_composition.scm": "9b56dc78823e7d35adb213744c0f8e95684074d6a12bb98c8c05755788be5b06", + "tests/test/lazy_evaluation.scm": "ea5912a53a35b65bf2bee9483dca90814b8baf52ae3ae50c2d4589f794dca54d", + "tests/test/list_operations.scm": "d7e70621afbbb787cf646e2089bccfbd3681fbb2e70325f3e50f56ccd30dbd9d", + "tests/test/map_operations.scm": "0cebddb3d4a50e4b6f6bfc0e8c12762d3f42609030aef077d8ab4bedda4718d9", + "tests/test/memoization.scm": "cf8202df22407dbba72ca7f660d8ab7cd4ca962b215c36cc491d5c1a474c160b", + "tests/test/mutual_recursion.scm": "a3dc23a96ec1cf05b0be4a4f0b6d3da9e2bad1aa3f7a83c7566f71f980a67fe2", + "tests/test/nested_defines.scm": "21faa0ccaf744cb8c9d283bbfd0c36e195c803c85f511e70a99267c0a85b6216", + "tests/test/oeis_sequences.scm": "d1f77161b367176b3dfab55a34466e511c602c7ae359e05f5a715627b93dca6a", + "tests/test/oeis_sequences2.scm": "5679ef0f99743114d8c06edcdcde1fea9474b71aeb57c2069f1ee33df1970bce", + "tests/test/oeis_sequences3.scm": "5dc8b44158b2648a6e9766f6bcdf21553a07bf487d79e64ebb418511ef3b8186", + "tests/test/recursive_structures.scm": "67838ff5bac26ea359139720adec1d5700176147b0062065c7def9715e8a2aeb", + "tests/test/test_read.scm": "5b9ee381c3cf45b2fc8c2e7b490857fb67f1ac78653b586d7f2848daf6853809", + "tests/test/variadic_functions.scm": "c9b847c50840c6977904e5033a07f6108cc3a54012b02d916e79025fb43e0f6e", + "tests/test/y_combinator.scm": "971144dcb2b04aaa3b34d9279459bd7d2e8256c585d17809465848c2589d36d8", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "fbb84e3c0c2f72ed7de90f54380f860052c16155e7d0c9076cd57c966e9be7ed" + }, + "prior_nanopy_work_budget_seconds": 2220, + "pi_agent_timeout_seconds": 2400 + }, + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/torch-pipeline-parallelism:20251031", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "297ddea07c230dfd246a8ab07337dd7abcd8e5e2a2b2e26590cb0a8af779839c", + "environment/Dockerfile": "c357ea41e4a4013c5c5980b8fef04af6a5ae250ccc6697cf3bad3127401a931f", + "instruction.md": "989bfcf0c54baf6c8cdc391743dbb2f9e858c5779dd85112536875d4d1b2dcd2", + "solution/solve.sh": "787198e4178d88b887697aefdd29819f29435ffdfe356545d92f58a30fb38b70", + "task.toml": "8b4f9f525c08afbbc4eca9168e46238a61ccb903486773e9ca4da64bb7ffe857", + "tests/test.sh": "cc9fd28b6590993fc37e4a99da771cbf04726c8c62fe308e802612241da69dea", + "tests/test_outputs.py": "23574246a407163a8e6581c84287cc2f1eb745efbaad2a317a5a28c39958c64f" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/torch-tensor-parallelism:20251031", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "4c1b3c3be30548f1f94c6f5503b1bc74b006af7f192dc4cd026f411df1a54604", + "environment/Dockerfile": "5f264ff0bf62a49aee3c6398447de5108df141e7a2ad35c6de8ef39fa62049c6", + "instruction.md": "13bde2ab165d72a311e2bf066d5d6c3c8761bf9d3dd39d1b23eda23bc546b910", + "solution/solve.sh": "d083b7cbf9dbb6cc2ff38c8f2acc150f0059c04aea42d6c2183aeaecc5a19385", + "task.toml": "72fd2a5500d1fe8795dcae76cccb1a29fb25894881a7475f2fa519c87d425a8d", + "tests/test.sh": "a7e6c9408b2401e4e7448371f7ddc72f3284caccfb56277d3cd4e98051868cdf", + "tests/test_outputs.py": "4ae4285e15fba1fcf13d7b7c0638ac8c2e2b0797c937abab1fb6b136fd508290" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/write-compressor", + "ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/write-compressor:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "not_configured", + "task_file_sha256": { + "README.md": "603015a28500a28f0838fea2c43a5168613746bfd07737920f4bae40b5a326f7", + "environment/Dockerfile": "9a583d29d46818f6ff2be7c6a0879d03343106e5376fd5a48eb5926b402bb93b", + "environment/data.txt": "0fc483d0fb31128dfa41ed11adca56c88688888189826ef9daeafd7559b4fa4e", + "environment/decomp.c": "7f8d9ef66943f912a217aa48f4ef5efc967b44c2b21d0f56df029b5ab413f62a", + "environment/main.rs": "fa53d6572fca95903e30d2d953f1e428bb8b63513ff218ebee9bc60a425ab389", + "instruction.md": "c4fc3e73e44deeac95f1cd4a34a2d87e853b22ceec2b2824767f5f087b470bb2", + "solution/solve.sh": "01550b7de7ab1320b1898eae055226bb7c73bee259c75102d2240c974a730409", + "task.toml": "6066a509066433c97c417fea68c4e3e73996f698dcdcbabe07718835eea355a6", + "tests/original-data.txt": "0fc483d0fb31128dfa41ed11adca56c88688888189826ef9daeafd7559b4fa4e", + "tests/original-decomp.c": "7f8d9ef66943f912a217aa48f4ef5efc967b44c2b21d0f56df029b5ab413f62a", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "809f486053e3ffbce852111dee1f63d80ab31297bd33172389e50278bed384e2" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + } + ], + "preflight": [ + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-01.json", + "config_sha256": "ff4f31f84f360a94c3efa4fffe569d56fdb803b0efb3cadeb577f068639d87e5", + "job_name": "dsh-dev34-install-pair-01-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-02.json", + "config_sha256": "e339a7afbb5af75556239a458d322f8e62c22b5f48c087315918001768b29a67", + "job_name": "dsh-dev34-install-pair-02-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-03.json", + "config_sha256": "e271aa10112d552815abdf154cc08571bf1b994f1025411f3653dc458f445c30", + "job_name": "dsh-dev34-install-pair-03-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-04.json", + "config_sha256": "e563ce2c6b51c7305db9e192fbb2928cacc7db0b6daf357ba142fcf6ad04e596", + "job_name": "dsh-dev34-install-pair-04-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-05.json", + "config_sha256": "581fa3b144a8a972b862ecef4593aa07dda2fcad1aca24f71d120a79e6f4f1dc", + "job_name": "dsh-dev34-install-pair-05-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-06.json", + "config_sha256": "b464fd2e0b96439026c364e799d28dac2c79fcacd6adff1799d88500f7762252", + "job_name": "dsh-dev34-install-pair-06-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-07.json", + "config_sha256": "c600144fb5e84f38e027ace5db42e8e129a52a389570dbf3be9d4c004647ff67", + "job_name": "dsh-dev34-install-pair-07-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-08.json", + "config_sha256": "9bd98339613ef303c53d2460c544bc888772fb8d746937d92e19960567c83f5e", + "job_name": "dsh-dev34-install-pair-08-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-09.json", + "config_sha256": "5ba07982f43ec0d6d5088e8869807509d78967e3a87120b9d2eb768b341709df", + "job_name": "dsh-dev34-install-pair-09-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-10.json", + "config_sha256": "ec06efb8808924f51bda72112238f644ff2fd0928bf64d7d62fbc068ec16650b", + "job_name": "dsh-dev34-install-pair-10-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-11.json", + "config_sha256": "b5b20a21de9867dcbd11ba54f7408252e6b6783b52ab81fdb09a0258fbb534af", + "job_name": "dsh-dev34-install-pair-11-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-12.json", + "config_sha256": "f228e9fa5263dfc533474a01eea066eea7d1f3cc5a65660f78619a8bf472be7b", + "job_name": "dsh-dev34-install-pair-12-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-13.json", + "config_sha256": "eadba0d82dc33942583a3d94260675615314703304e9a61651e853e5e1703cab", + "job_name": "dsh-dev34-install-pair-13-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-14.json", + "config_sha256": "c7ee6206106c1559c69ebb8533fb2338d17d4c2134d24baecc619676b3f8c433", + "job_name": "dsh-dev34-install-pair-14-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-15.json", + "config_sha256": "9c05600fb4963450eeec633ffdf875aa64894b202d8223b094ba3a1f7203207e", + "job_name": "dsh-dev34-install-pair-15-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-16.json", + "config_sha256": "5574fa316a3d056b792e15e35cab9d176143c29a5b23e54072a3d1bdf236d294", + "job_name": "dsh-dev34-install-pair-16-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-17.json", + "config_sha256": "fba8e1ce19de42ddc4ea28a5941281f6a9c657a2eab8a4bcafbd58004a948463", + "job_name": "dsh-dev34-install-pair-17-20261030" + } + ], + "queue": [ + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/adaptive-rejection-sampler.json", + "config_sha256": "207be577c3540c83b5a826a4b3af05f1d5928ba68c2259e21bfb03bb5157212e", + "task": "terminal-bench/adaptive-rejection-sampler", + "repeat": 1, + "job_name": "dsh-dev34-adaptive-rejection-sampler-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-cython-ext.json", + "config_sha256": "5f8d145c831d126078c5b89918784d3837dbe1973874ccfb87155819a5fc8c69", + "task": "terminal-bench/build-cython-ext", + "repeat": 1, + "job_name": "dsh-dev34-build-cython-ext-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cancel-async-tasks.json", + "config_sha256": "db8fab736b87703557ddc062c6db2bd8d50d973b0a28ac3c1e89d4f2b9c23fb3", + "task": "terminal-bench/cancel-async-tasks", + "repeat": 1, + "job_name": "dsh-dev34-cancel-async-tasks-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/cobol-modernization.json", + "config_sha256": "bc8f7cd7a06b82b1af6d212c9085f14cd60675338ef1f4793bb3b008743b0e0a", + "task": "terminal-bench/cobol-modernization", + "repeat": 1, + "job_name": "dsh-dev34-cobol-modernization-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/db-wal-recovery.json", + "config_sha256": "58f5cf40b26027aa650be04c5ac0b0a070a4ebb964004af7b39d44376823fc8a", + "task": "terminal-bench/db-wal-recovery", + "repeat": 1, + "job_name": "dsh-dev34-db-wal-recovery-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/fix-code-vulnerability.json", + "config_sha256": "63ab5f4f94c86c0290437818372360563ff64859aa28d33c21d557fa46b51cb4", + "task": "terminal-bench/fix-code-vulnerability", + "repeat": 1, + "job_name": "dsh-dev34-fix-code-vulnerability-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/git-multibranch.json", + "config_sha256": "f449471f210950b8e9d85b65237091fe2d20956d7acce31909ef91c1e77f9147", + "task": "terminal-bench/git-multibranch", + "repeat": 1, + "job_name": "dsh-dev34-git-multibranch-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/kv-store-grpc.json", + "config_sha256": "4b7e7387c890ec4ec066f7fd479335b3f546da7d2378b1100408a3a7b862cfe3", + "task": "terminal-bench/kv-store-grpc", + "repeat": 1, + "job_name": "dsh-dev34-kv-store-grpc-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/log-summary-date-ranges.json", + "config_sha256": "bc4ad9dfd2c0db98d5f6fed56c44742550dcc175f7b167e0e3fb81dcebd48106", + "task": "terminal-bench/log-summary-date-ranges", + "repeat": 1, + "job_name": "dsh-dev34-log-summary-date-ranges-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/merge-diff-arc-agi-task.json", + "config_sha256": "372ef6ac5628653c889b69463d760987f77f3f4221058d6f129c467e4dfd3ece", + "task": "terminal-bench/merge-diff-arc-agi-task", + "repeat": 1, + "job_name": "dsh-dev34-merge-diff-arc-agi-task-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-extraction-relu-logits.json", + "config_sha256": "fbb49d9d7454e34dcbc9f463047cb09c7596aa76d2040aec41fb3cdb4e100442", + "task": "terminal-bench/model-extraction-relu-logits", + "repeat": 1, + "job_name": "dsh-dev34-model-extraction-relu-logits-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/multi-source-data-merger.json", + "config_sha256": "f0f7e4680dbe4eb2e3b351a85aec1663f94d4494cf347d6b09fa57b6a04794cc", + "task": "terminal-bench/multi-source-data-merger", + "repeat": 1, + "job_name": "dsh-dev34-multi-source-data-merger-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/nginx-request-logging.json", + "config_sha256": "c582348272d160d158844bef121ea44511373a4d7f209e576930635176fff98a", + "task": "terminal-bench/nginx-request-logging", + "repeat": 1, + "job_name": "dsh-dev34-nginx-request-logging-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/openssl-selfsigned-cert.json", + "config_sha256": "d2bedba396484ce12b21116e477622b44711799aede75d3e497c466ce114bf92", + "task": "terminal-bench/openssl-selfsigned-cert", + "repeat": 1, + "job_name": "dsh-dev34-openssl-selfsigned-cert-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pypi-server.json", + "config_sha256": "0fb5103a50f99d6c67d443d4bb4c0e4ba1ff44027e649f0fef6051a881e0db90", + "task": "terminal-bench/pypi-server", + "repeat": 1, + "job_name": "dsh-dev34-pypi-server-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-cli.json", + "config_sha256": "6f8422d6df51425da8d2788f5cfb0133e6d07af01470e5ac8800c11fa9622605", + "task": "terminal-bench/pytorch-model-cli", + "repeat": 1, + "job_name": "dsh-dev34-pytorch-model-cli-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-recovery.json", + "config_sha256": "d69b11ca524da6c967d7bcee271b2eb0d241df7be14859ed01ab399082e04471", + "task": "terminal-bench/pytorch-model-recovery", + "repeat": 1, + "job_name": "dsh-dev34-pytorch-model-recovery-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/qemu-alpine-ssh.json", + "config_sha256": "b7bf1993548ba6f5c16e7a0dd509baf4097fc7af0249ab8656c6dd2b13d26c88", + "task": "terminal-bench/qemu-alpine-ssh", + "repeat": 1, + "job_name": "dsh-dev34-qemu-alpine-ssh-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/query-optimize.json", + "config_sha256": "c2c2a93e6ea320578b05fb7927ca558a47a8867cbf2ad2a49ef8841f471b1995", + "task": "terminal-bench/query-optimize", + "repeat": 1, + "job_name": "dsh-dev34-query-optimize-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-log.json", + "config_sha256": "70a4847f93da146778a72490bd0f3897bd39838737552f55c5799372e205960f", + "task": "terminal-bench/regex-log", + "repeat": 1, + "job_name": "dsh-dev34-regex-log-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-pipeline-parallelism.json", + "config_sha256": "8e9ec1292449bb58d666e3e2f5c61cef796e8b44ce8fc55e9127cd51a2be6634", + "task": "terminal-bench/torch-pipeline-parallelism", + "repeat": 1, + "job_name": "dsh-dev34-torch-pipeline-parallelism-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-tensor-parallelism.json", + "config_sha256": "323f1ece73ca6c3f1ed42ff825db05a1a3ff23d05a2ea66008ca781546201a53", + "task": "terminal-bench/torch-tensor-parallelism", + "repeat": 1, + "job_name": "dsh-dev34-torch-tensor-parallelism-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/write-compressor.json", + "config_sha256": "1b1be0e871bb8fbfdf75a1161032ead5445c38e5d505e1e68d15f9b404ce6167", + "task": "terminal-bench/write-compressor", + "repeat": 1, + "job_name": "dsh-dev34-write-compressor-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/constraints-scheduling.json", + "config_sha256": "cf0c88f9e35ab2097400cc8289719bdcda4aa5ce53adaa517eef8d229b65dd78", + "task": "terminal-bench/constraints-scheduling", + "repeat": 1, + "job_name": "dsh-dev34-constraints-scheduling-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/custom-memory-heap-crash.json", + "config_sha256": "64a026fc8dd82b2c44756d5d8d14f868381f534579455b13c3bcc8cf7e5f8b9f", + "task": "terminal-bench/custom-memory-heap-crash", + "repeat": 1, + "job_name": "dsh-dev34-custom-memory-heap-crash-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dna-assembly.json", + "config_sha256": "ad2a7ef2ec1a82eae0ad16abd909116bd0e96e1ccce4b27de46caeceea2efa53", + "task": "terminal-bench/dna-assembly", + "repeat": 1, + "job_name": "dsh-dev34-dna-assembly-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/llm-inference-batching-scheduler.json", + "config_sha256": "2a3e31aaf3c3b6fcb1f9f509a478faadc9ab4bda00b483d4678999016093ec8e", + "task": "terminal-bench/llm-inference-batching-scheduler", + "repeat": 1, + "job_name": "dsh-dev34-llm-inference-batching-scheduler-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/path-tracing.json", + "config_sha256": "be8db081812b9e9f838b5e08286eb5e3a37e06cc4a1814416fe1489787495527", + "task": "terminal-bench/path-tracing", + "repeat": 1, + "job_name": "dsh-dev34-path-tracing-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/schemelike-metacircular-eval.json", + "config_sha256": "0d168659d04ea65ebad140b574e77c4722c0be11987d31c9ce492a68cb3463b7", + "task": "terminal-bench/schemelike-metacircular-eval", + "repeat": 1, + "job_name": "dsh-dev34-schemelike-metacircular-eval-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/caffe-cifar-10.json", + "config_sha256": "29fb6066d4b2a3167226e4e29138f4168d67925d21ba1cf838e276dda6e0f636", + "task": "terminal-bench/caffe-cifar-10", + "repeat": 1, + "job_name": "dsh-dev34-caffe-cifar-10-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/circuit-fibsqrt.json", + "config_sha256": "b3c54e46b9f9829730dc5f3dcdd1f45df8bd5147b1730f3267a2e5da1cc20ce6", + "task": "terminal-bench/circuit-fibsqrt", + "repeat": 1, + "job_name": "dsh-dev34-circuit-fibsqrt-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/mteb-leaderboard.json", + "config_sha256": "26d531025a116f70d795665698be401478331be03b3be560989e38ca57eafb06", + "task": "terminal-bench/mteb-leaderboard", + "repeat": 1, + "job_name": "dsh-dev34-mteb-leaderboard-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-chess.json", + "config_sha256": "83fc1dcc75d7bd601e12bf4fa93a004ffd8dced7c34f2d88456131a9e3a81716", + "task": "terminal-bench/regex-chess", + "repeat": 1, + "job_name": "dsh-dev34-regex-chess-20261030" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/build-pov-ray.json", + "config_sha256": "d547bcbdb683a6f27170c9019f3d25de4b671ec693236192ff3178a551af9b73", + "task": "terminal-bench/build-pov-ray", + "repeat": 1, + "job_name": "dsh-dev34-build-pov-ray-20261030" + } + ], + "additional_file_sha256": { + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/selection.json": "4572b1cc432a16b9c5d1dd554c8d2189ae80414f386a90bf03829bc4d0c3da85", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/image-digests.json": "7f21855951ff5fff62cb16e686d93e4aae726be117dad049641c67956ab00d46", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-release-manifest.json": "77dfda40e45b870a8ce7f07a1e058d416d94a04ab3d27c2f21bc370f22151394", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/apt-cache.conf": "d0e1ffb6dbd02c54978ba2df386326311a59f6d3774990180f22bcc6adfaedbc", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json": "acdb4d3f502cf062395f708e4684a1b308db8ec2133ea6a74eee04d272794573", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/deepseek-distribution.json": "34410bdce15ec3b417105eb7f9f6e01a14c47f1bfd0ffe16d76338e7efa51dd2", + "benchmarks/harbor/scripts/run_comparison.py": "228dbff33c0ec8a2359e612da632ef543dcac4273db2f9a7d0573a24e856fbae", + "benchmarks/harbor/scripts/apt_index_proxy.py": "bc15f7adf1d7968d627ceca4683ab86742a7188a1b1955ba1fef37ba48e8a936", + "benchmarks/harbor/scripts/cache_image.py": "11ddde3fe86a6e4bfdefe7c3578794df98d237f701e40f219d76a541911c9f0c", + "benchmarks/harbor/scripts/uv_release_cache.py": "993cc35b1f3a9e8e171b662ab6d4524c5fe059f0d1f0e275a5bb1930111e0ff5", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json": "118edc62f44e571ee2df5bbd41f6ab6fffc78a22908869dcaeffd11c6c9a8fbf" + }, + "batch_images": { + "manifest": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/image-digests.json", + "storage_path": "/var/lib/docker", + "minimum_free_disk_bytes": 16106127360, + "cache_dir": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/images" + }, + "model_policy": { + "model": "deepseek/deepseek-flash", + "provider": "deepseek-official", + "api": "anthropic-messages", + "base_url": "https://api.deepseek.com/anthropic", + "reasoning_effort": "high", + "profile": "sdk-minimal", + "temperature": "unset", + "sdk_max_tokens": "unset", + "upstream_default_max_tokens": 256000, + "alias": "Official documentation checked 2026-10-30 maps the mutable deepseek-flash alias to DeepSeek-V4.1-Flash-1017." + }, + "budget_policy": "Original task-native timeouts/resources; 1800-second setup. Upstream sdk-minimal prompts, tools, compaction, retries, and generation defaults. No injected turn limit, internal time budget, or finalization reminders. Pi used OpenAI-compatible transport with 384000-token model metadata; this is not a matched-generation-budget causal harness comparison.", + "verifier_policy": "Original official tests and native per-attempt timeout; retry only verifier timeouts, at most three total attempts against unchanged agent output. No dependency-error recovery. Require confirmed completion or cancellation before verification.", + "preflight_policy": "Fresh install-only checks for all 34 tasks before any model trial. QEMU has the registered qemu-alpine-ssh-v1 profile; the other 33 explicitly report not_configured. SDK installation success does not certify unsupported verifier dependencies.", + "information_policy": "All 34 tasks are the existing development pool with prior exposure. No outcome-based task replacement. Review only operational metadata while running; freeze all new outputs before inspecting new trajectories or hidden tests. Keep the remaining 55 tasks unopened.", + "runtime_environment": "UV_PROJECT_ENVIRONMENT=/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/dsh-harbor-venv", + "status_at_registration": "No model attempts. Official sdk-minimal keyless smoke and corrected WAL container installation smoke passed. All 34 registered preflights remain required.", + "historical_pi_results": "benchmarks/harbor/results/pi-deepseek-flash-dev34-20261024.json" +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/fix-code-vulnerability.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/fix-code-vulnerability.json new file mode 100644 index 0000000..ef037a8 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/fix-code-vulnerability.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/fix-code-vulnerability", + "ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/git-multibranch.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/git-multibranch.json new file mode 100644 index 0000000..5f99e93 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/git-multibranch.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/git-multibranch", + "ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/image-digests.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/image-digests.json new file mode 100644 index 0000000..c5661be --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/image-digests.json @@ -0,0 +1,312 @@ +{ + "observed_at": "2026-10-24T14:07:09.186521+00:00", + "images": [ + { + "task": "terminal-bench/adaptive-rejection-sampler", + "image": "alexgshaw/adaptive-rejection-sampler:20251031", + "image_id": "sha256:985bed39e8448bc95f138b5f766b59003a95c4be171c96af8e76448338f3dd8f", + "repo_digests": [ + "alexgshaw/adaptive-rejection-sampler@sha256:985bed39e8448bc95f138b5f766b59003a95c4be171c96af8e76448338f3dd8f" + ], + "size_bytes": 29731756 + }, + { + "task": "terminal-bench/build-cython-ext", + "image": "alexgshaw/build-cython-ext:20251031", + "image_id": "sha256:3612a38fadb89a96f74a1a951fb0b0af734198fd160571eeaba6401593234594", + "repo_digests": [ + "alexgshaw/build-cython-ext@sha256:3612a38fadb89a96f74a1a951fb0b0af734198fd160571eeaba6401593234594" + ], + "size_bytes": 303222492 + }, + { + "task": "terminal-bench/cancel-async-tasks", + "image": "alexgshaw/cancel-async-tasks:20251031", + "image_id": "sha256:84c7fae6b256dcc56a350790e2a9715eefc7dad662a9d8e8a472363aa71ef18d", + "repo_digests": [ + "alexgshaw/cancel-async-tasks@sha256:84c7fae6b256dcc56a350790e2a9715eefc7dad662a9d8e8a472363aa71ef18d" + ], + "size_bytes": 44171676 + }, + { + "task": "terminal-bench/cobol-modernization", + "image": "alexgshaw/cobol-modernization:20251031", + "image_id": "sha256:593ab9df3d83f771e927888b2d9436b380cdf7ec4d0af2a9f0802d541edebba0", + "repo_digests": [ + "alexgshaw/cobol-modernization@sha256:593ab9df3d83f771e927888b2d9436b380cdf7ec4d0af2a9f0802d541edebba0" + ], + "size_bytes": 162020401 + }, + { + "task": "terminal-bench/db-wal-recovery", + "image": "alexgshaw/db-wal-recovery:20251031", + "image_id": "sha256:0e33ea5ec823975d1bd6c3778395c9f94251dd88f571146057bff6adb7e4594e", + "repo_digests": [ + "alexgshaw/db-wal-recovery@sha256:0e33ea5ec823975d1bd6c3778395c9f94251dd88f571146057bff6adb7e4594e" + ], + "size_bytes": 175276145 + }, + { + "task": "terminal-bench/fix-code-vulnerability", + "image": "alexgshaw/fix-code-vulnerability:20251031", + "image_id": "sha256:cac325252991f823713b2d0441502972901dd782bd67f66c03d9b1e410dac5c0", + "repo_digests": [ + "alexgshaw/fix-code-vulnerability@sha256:cac325252991f823713b2d0441502972901dd782bd67f66c03d9b1e410dac5c0" + ], + "size_bytes": 190696453 + }, + { + "task": "terminal-bench/git-multibranch", + "image": "alexgshaw/git-multibranch:20251031", + "image_id": "sha256:5bc449921c138295f6ddd8dca2977319fee2ce523811dbe56cec4ca663385fb5", + "repo_digests": [ + "alexgshaw/git-multibranch@sha256:5bc449921c138295f6ddd8dca2977319fee2ce523811dbe56cec4ca663385fb5" + ], + "size_bytes": 222134454 + }, + { + "task": "terminal-bench/kv-store-grpc", + "image": "alexgshaw/kv-store-grpc:20251031", + "image_id": "sha256:3399400800dcb207634daa42bc1b052e831e285cc9d221eea66c47bc0fc79791", + "repo_digests": [ + "alexgshaw/kv-store-grpc@sha256:3399400800dcb207634daa42bc1b052e831e285cc9d221eea66c47bc0fc79791" + ], + "size_bytes": 44171671 + }, + { + "task": "terminal-bench/log-summary-date-ranges", + "image": "alexgshaw/log-summary-date-ranges:20251031", + "image_id": "sha256:cbeb6ba905c2fec294f16cd5e16e3ea7f2e04d38ac2484d51a11de262aa7dc51", + "repo_digests": [ + "alexgshaw/log-summary-date-ranges@sha256:cbeb6ba905c2fec294f16cd5e16e3ea7f2e04d38ac2484d51a11de262aa7dc51" + ], + "size_bytes": 44866158 + }, + { + "task": "terminal-bench/merge-diff-arc-agi-task", + "image": "alexgshaw/merge-diff-arc-agi-task:20251031", + "image_id": "sha256:bfc2a235f2ceea64ffb1ffc5529c2acebf570bff61f3668f0e5669ef570c17d4", + "repo_digests": [ + "alexgshaw/merge-diff-arc-agi-task@sha256:bfc2a235f2ceea64ffb1ffc5529c2acebf570bff61f3668f0e5669ef570c17d4" + ], + "size_bytes": 29728285 + }, + { + "task": "terminal-bench/model-extraction-relu-logits", + "image": "alexgshaw/model-extraction-relu-logits:20251031", + "image_id": "sha256:52fe1f089f38650f0dc22d7531ee6e01ebd526de1b349aaa2ecb331dca9fabce", + "repo_digests": [ + "alexgshaw/model-extraction-relu-logits@sha256:52fe1f089f38650f0dc22d7531ee6e01ebd526de1b349aaa2ecb331dca9fabce" + ], + "size_bytes": 83698971 + }, + { + "task": "terminal-bench/multi-source-data-merger", + "image": "alexgshaw/multi-source-data-merger:20251031", + "image_id": "sha256:8b32782078ff7383a1b4e5d3cecca8ce287e30f50bb4c0e5db009a18064e666e", + "repo_digests": [ + "alexgshaw/multi-source-data-merger@sha256:8b32782078ff7383a1b4e5d3cecca8ce287e30f50bb4c0e5db009a18064e666e" + ], + "size_bytes": 220880128 + }, + { + "task": "terminal-bench/nginx-request-logging", + "image": "alexgshaw/nginx-request-logging:20251031", + "image_id": "sha256:693e64431a0dbf45952e8b91a298d54fd60247e501c6623fb8fa3ebc07fc3d3d", + "repo_digests": [ + "alexgshaw/nginx-request-logging@sha256:693e64431a0dbf45952e8b91a298d54fd60247e501c6623fb8fa3ebc07fc3d3d" + ], + "size_bytes": 67285778 + }, + { + "task": "terminal-bench/openssl-selfsigned-cert", + "image": "alexgshaw/openssl-selfsigned-cert:20251031", + "image_id": "sha256:4c948a4e630af2435ae0a19108fc0814a946ac2fa29a512469e0fc77b38c8c12", + "repo_digests": [ + "alexgshaw/openssl-selfsigned-cert@sha256:4c948a4e630af2435ae0a19108fc0814a946ac2fa29a512469e0fc77b38c8c12" + ], + "size_bytes": 44173158 + }, + { + "task": "terminal-bench/pypi-server", + "image": "alexgshaw/pypi-server:20251031", + "image_id": "sha256:d18bb30f47c7dcaa3acdbc31ac7413c98eadca2ea16540bbd380f2f063b95276", + "repo_digests": [ + "alexgshaw/pypi-server@sha256:d18bb30f47c7dcaa3acdbc31ac7413c98eadca2ea16540bbd380f2f063b95276" + ], + "size_bytes": 83726049 + }, + { + "task": "terminal-bench/pytorch-model-cli", + "image": "alexgshaw/pytorch-model-cli:20251031", + "image_id": "sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "repo_digests": [ + "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67" + ], + "size_bytes": 44349611 + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "image": "alexgshaw/pytorch-model-recovery:20260430", + "image_id": "sha256:7c2bb52851cef25bdc3939e473496d25fbf11fd0cb7ab58bcf3cefcb751cdc0e", + "repo_digests": [ + "alexgshaw/pytorch-model-recovery@sha256:7c2bb52851cef25bdc3939e473496d25fbf11fd0cb7ab58bcf3cefcb751cdc0e" + ], + "size_bytes": 269661811 + }, + { + "task": "terminal-bench/qemu-alpine-ssh", + "image": "alexgshaw/qemu-alpine-ssh:20251031", + "image_id": "sha256:541e8214a8c4faa7b556531b899d45da21048fa2ca6dd328d8cb6c4cffa8e964", + "repo_digests": [ + "alexgshaw/qemu-alpine-ssh@sha256:541e8214a8c4faa7b556531b899d45da21048fa2ca6dd328d8cb6c4cffa8e964" + ], + "size_bytes": 1308553944 + }, + { + "task": "terminal-bench/query-optimize", + "image": "alexgshaw/query-optimize:20251031", + "image_id": "sha256:28de7da87004eab5b29bd5eaf1823579bc5b73ccea4b80349a5623185770da14", + "repo_digests": [ + "alexgshaw/query-optimize@sha256:28de7da87004eab5b29bd5eaf1823579bc5b73ccea4b80349a5623185770da14" + ], + "size_bytes": 120494963 + }, + { + "task": "terminal-bench/regex-log", + "image": "alexgshaw/regex-log:20251031", + "image_id": "sha256:90101b2e815323a8da20528a1439bebc407eb9761c9c68a3d557730856c878e9", + "repo_digests": [ + "alexgshaw/regex-log@sha256:90101b2e815323a8da20528a1439bebc407eb9761c9c68a3d557730856c878e9" + ], + "size_bytes": 29725514 + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "image": "alexgshaw/torch-pipeline-parallelism:20251031", + "image_id": "sha256:3cb7b39d86f1704b20e668851da0b08b34b3270c01d583f136d65190984a8d1b", + "repo_digests": [ + "alexgshaw/torch-pipeline-parallelism@sha256:3cb7b39d86f1704b20e668851da0b08b34b3270c01d583f136d65190984a8d1b" + ], + "size_bytes": 29725531 + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "image": "alexgshaw/torch-tensor-parallelism:20251031", + "image_id": "sha256:c50ac169e11465f41b49749fe87e4487f24d31a0c39c0906cd3242cc5499c690", + "repo_digests": [ + "alexgshaw/torch-tensor-parallelism@sha256:c50ac169e11465f41b49749fe87e4487f24d31a0c39c0906cd3242cc5499c690" + ], + "size_bytes": 29725529 + }, + { + "task": "terminal-bench/write-compressor", + "image": "alexgshaw/write-compressor:20251031", + "image_id": "sha256:3618e1f8a997b437c09cd3dbaac736705809c05f6f12f584b65722175970ebe1", + "repo_digests": [ + "alexgshaw/write-compressor@sha256:3618e1f8a997b437c09cd3dbaac736705809c05f6f12f584b65722175970ebe1" + ], + "size_bytes": 291927726 + }, + { + "task": "terminal-bench/constraints-scheduling", + "image": "alexgshaw/constraints-scheduling:20251031", + "image_id": "sha256:567ce5a189f8d11ac461790876e934cc7af38391baf89a78f95f4dafc1fec3b0", + "repo_digests": [ + "alexgshaw/constraints-scheduling@sha256:567ce5a189f8d11ac461790876e934cc7af38391baf89a78f95f4dafc1fec3b0" + ], + "size_bytes": 49852952 + }, + { + "task": "terminal-bench/custom-memory-heap-crash", + "image": "alexgshaw/custom-memory-heap-crash:20251031", + "image_id": "sha256:1895d7d6240d5fb2ec774d6bfdd1057d170f8d199c15c582e4585d3c63802557", + "repo_digests": [ + "alexgshaw/custom-memory-heap-crash@sha256:1895d7d6240d5fb2ec774d6bfdd1057d170f8d199c15c582e4585d3c63802557" + ], + "size_bytes": 1267999268 + }, + { + "task": "terminal-bench/dna-assembly", + "image": "alexgshaw/dna-assembly:20251031", + "image_id": "sha256:d1adf6835f1dd91205ba70e452c699d0aea601010038e5617f370716efb50569", + "repo_digests": [ + "alexgshaw/dna-assembly@sha256:d1adf6835f1dd91205ba70e452c699d0aea601010038e5617f370716efb50569" + ], + "size_bytes": 29727501 + }, + { + "task": "terminal-bench/llm-inference-batching-scheduler", + "image": "alexgshaw/llm-inference-batching-scheduler:20251031", + "image_id": "sha256:19e79aa49be4e55dccca1dde966515ebe6852314b6010bcb9230afb48b996774", + "repo_digests": [ + "alexgshaw/llm-inference-batching-scheduler@sha256:19e79aa49be4e55dccca1dde966515ebe6852314b6010bcb9230afb48b996774" + ], + "size_bytes": 44188631 + }, + { + "task": "terminal-bench/path-tracing", + "image": "alexgshaw/path-tracing:20251031", + "image_id": "sha256:4b44725102cd627dfa895191a5a30d16dae3c5623bb43da6459f79135db6b482", + "repo_digests": [ + "alexgshaw/path-tracing@sha256:4b44725102cd627dfa895191a5a30d16dae3c5623bb43da6459f79135db6b482" + ], + "size_bytes": 399390468 + }, + { + "task": "terminal-bench/schemelike-metacircular-eval", + "image": "alexgshaw/schemelike-metacircular-eval:20251031", + "image_id": "sha256:b485be38bf80d56c1a4cd9da95bbf7ec129dc03f6786b86c0525518a96f6d1db", + "repo_digests": [ + "alexgshaw/schemelike-metacircular-eval@sha256:b485be38bf80d56c1a4cd9da95bbf7ec129dc03f6786b86c0525518a96f6d1db" + ], + "size_bytes": 44188336 + }, + { + "task": "terminal-bench/caffe-cifar-10", + "image": "alexgshaw/caffe-cifar-10:20260403", + "image_id": "sha256:929a6d631b592c82e0d0f5d4a65cf03c946dc554cfc9bb09916e82076e7d3215", + "repo_digests": [ + "alexgshaw/caffe-cifar-10@sha256:929a6d631b592c82e0d0f5d4a65cf03c946dc554cfc9bb09916e82076e7d3215" + ], + "size_bytes": 213171982 + }, + { + "task": "terminal-bench/circuit-fibsqrt", + "image": "alexgshaw/circuit-fibsqrt:20251031", + "image_id": "sha256:29783439f529eaed2145592f15af7a2281161528860392a382c825489030ae3a", + "repo_digests": [ + "alexgshaw/circuit-fibsqrt@sha256:29783439f529eaed2145592f15af7a2281161528860392a382c825489030ae3a" + ], + "size_bytes": 149136738 + }, + { + "task": "terminal-bench/mteb-leaderboard", + "image": "alexgshaw/mteb-leaderboard:20260430", + "image_id": "sha256:a8538f4882ff132e0a20a75be3fffc07164318a7425ad19a6b4bf5268107feb4", + "repo_digests": [ + "alexgshaw/mteb-leaderboard@sha256:a8538f4882ff132e0a20a75be3fffc07164318a7425ad19a6b4bf5268107feb4" + ], + "size_bytes": 663825427 + }, + { + "task": "terminal-bench/regex-chess", + "image": "alexgshaw/regex-chess:20251031", + "image_id": "sha256:e3b6d61d2d3930fbbdc5f7bf091a7f8b7f6125e1951ef8a7dbe165b1ef62942c", + "repo_digests": [ + "alexgshaw/regex-chess@sha256:e3b6d61d2d3930fbbdc5f7bf091a7f8b7f6125e1951ef8a7dbe165b1ef62942c" + ], + "size_bytes": 55625584 + }, + { + "task": "terminal-bench/build-pov-ray", + "image": "alexgshaw/build-pov-ray:20251031", + "image_id": "sha256:874baf49341e2853a15a9184e308c7eb632e97f434359fcff2891c6e5ca1b39b", + "repo_digests": [ + "alexgshaw/build-pov-ray@sha256:874baf49341e2853a15a9184e308c7eb632e97f434359fcff2891c6e5ca1b39b" + ], + "size_bytes": 163066497 + } + ], + "matches_prior_registration": true +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-01.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-01.json new file mode 100644 index 0000000..24f5ad9 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-01.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/adaptive-rejection-sampler", + "ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de" + }, + { + "name": "terminal-bench/build-cython-ext", + "ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-02.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-02.json new file mode 100644 index 0000000..3ea9207 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-02.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/cancel-async-tasks", + "ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1" + }, + { + "name": "terminal-bench/cobol-modernization", + "ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-03.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-03.json new file mode 100644 index 0000000..e44608b --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-03.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/db-wal-recovery", + "ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4" + }, + { + "name": "terminal-bench/fix-code-vulnerability", + "ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-04.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-04.json new file mode 100644 index 0000000..aebc3a7 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-04.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/git-multibranch", + "ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa" + }, + { + "name": "terminal-bench/kv-store-grpc", + "ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-05.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-05.json new file mode 100644 index 0000000..0a68ef1 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-05.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/log-summary-date-ranges", + "ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6" + }, + { + "name": "terminal-bench/merge-diff-arc-agi-task", + "ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-06.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-06.json new file mode 100644 index 0000000..a37f241 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-06.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/model-extraction-relu-logits", + "ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17" + }, + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-07.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-07.json new file mode 100644 index 0000000..41f01ab --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-07.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/nginx-request-logging", + "ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097" + }, + { + "name": "terminal-bench/openssl-selfsigned-cert", + "ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-08.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-08.json new file mode 100644 index 0000000..fa84e92 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-08.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pypi-server", + "ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1" + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-09.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-09.json new file mode 100644 index 0000000..547c78f --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-09.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + }, + { + "name": "terminal-bench/qemu-alpine-ssh", + "ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-10.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-10.json new file mode 100644 index 0000000..9b8cf6d --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-10.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + }, + { + "name": "terminal-bench/regex-log", + "ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-11.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-11.json new file mode 100644 index 0000000..d629a7f --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-11.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-12.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-12.json new file mode 100644 index 0000000..5d4a1b5 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-12.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/write-compressor", + "ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e" + }, + { + "name": "terminal-bench/constraints-scheduling", + "ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-13.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-13.json new file mode 100644 index 0000000..ad23846 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-13.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/custom-memory-heap-crash", + "ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e" + }, + { + "name": "terminal-bench/dna-assembly", + "ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-14.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-14.json new file mode 100644 index 0000000..c7c1697 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-14.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/llm-inference-batching-scheduler", + "ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67" + }, + { + "name": "terminal-bench/path-tracing", + "ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-15.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-15.json new file mode 100644 index 0000000..f7f51be --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-15.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/schemelike-metacircular-eval", + "ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6" + }, + { + "name": "terminal-bench/caffe-cifar-10", + "ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-16.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-16.json new file mode 100644 index 0000000..c8b4446 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-16.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/circuit-fibsqrt", + "ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3" + }, + { + "name": "terminal-bench/mteb-leaderboard", + "ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-17.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-17.json new file mode 100644 index 0000000..ca4842a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/install-pair-17.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/regex-chess", + "ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78" + }, + { + "name": "terminal-bench/build-pov-ray", + "ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/kv-store-grpc.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/kv-store-grpc.json new file mode 100644 index 0000000..e50031d --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/kv-store-grpc.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/kv-store-grpc", + "ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/llm-inference-batching-scheduler.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/llm-inference-batching-scheduler.json new file mode 100644 index 0000000..34ebefa --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/llm-inference-batching-scheduler.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/llm-inference-batching-scheduler", + "ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/log-summary-date-ranges.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/log-summary-date-ranges.json new file mode 100644 index 0000000..567f910 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/log-summary-date-ranges.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/log-summary-date-ranges", + "ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/merge-diff-arc-agi-task.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/merge-diff-arc-agi-task.json new file mode 100644 index 0000000..7ceefdb --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/merge-diff-arc-agi-task.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/merge-diff-arc-agi-task", + "ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-extraction-relu-logits.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-extraction-relu-logits.json new file mode 100644 index 0000000..e15fc59 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-extraction-relu-logits.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/model-extraction-relu-logits", + "ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/mteb-leaderboard.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/mteb-leaderboard.json new file mode 100644 index 0000000..4ae245f --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/mteb-leaderboard.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/mteb-leaderboard", + "ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/multi-source-data-merger.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/multi-source-data-merger.json new file mode 100644 index 0000000..4ab1dd8 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/multi-source-data-merger.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/nginx-request-logging.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/nginx-request-logging.json new file mode 100644 index 0000000..95d10a6 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/nginx-request-logging.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/nginx-request-logging", + "ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/openssl-selfsigned-cert.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/openssl-selfsigned-cert.json new file mode 100644 index 0000000..f8f201b --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/openssl-selfsigned-cert.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/openssl-selfsigned-cert", + "ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/path-tracing.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/path-tracing.json new file mode 100644 index 0000000..26b0d51 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/path-tracing.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/path-tracing", + "ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pypi-server.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pypi-server.json new file mode 100644 index 0000000..0ea450b --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pypi-server.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pypi-server", + "ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-cli.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-cli.json new file mode 100644 index 0000000..68a3f5f --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-cli.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-recovery.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-recovery.json new file mode 100644 index 0000000..9c77389 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/pytorch-model-recovery.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/qemu-alpine-ssh.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/qemu-alpine-ssh.json new file mode 100644 index 0000000..ea98f41 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/qemu-alpine-ssh.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/qemu-alpine-ssh", + "ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/query-optimize.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/query-optimize.json new file mode 100644 index 0000000..18a3a3a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/query-optimize.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-chess.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-chess.json new file mode 100644 index 0000000..22ccb6f --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-chess.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/regex-chess", + "ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-log.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-log.json new file mode 100644 index 0000000..c45db9e --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/regex-log.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/regex-log", + "ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/schemelike-metacircular-eval.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/schemelike-metacircular-eval.json new file mode 100644 index 0000000..eb99c4a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/schemelike-metacircular-eval.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/schemelike-metacircular-eval", + "ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/selection.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/selection.json new file mode 100644 index 0000000..e4bf64e --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/selection.json @@ -0,0 +1,150 @@ +{ + "schema_version": 1, + "selected_at": "2026-10-12T13:52:35.779627+00:00", + "source_ref": "2eb99b608dfb39464d17d109a4a8f11afda607a9", + "dataset_ref": "sha256:7d7bdc1cbedad549fc1140404bd4dc45e5fd0ea7c4186773687d177ad3a0699a", + "catalog_sha256": "7fc7d95a2077ede76c4f47890a09de93c2fe558252a122ef830288ad293ceaaa", + "selection_method": "Entire existing 34-task development pool, explicitly requested by the user. No selection by latest score and no outcome-based replacements.", + "prior_selection_file": "../dev12-baseline-20261011/selection.json", + "prior_selection_sha256": "cd51acea6592c01ce9ec8cde34cd46a4965a4164950194f2de7e8ca54649bd29", + "tasks": [ + { + "name": "terminal-bench/adaptive-rejection-sampler", + "ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de" + }, + { + "name": "terminal-bench/build-cython-ext", + "ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12" + }, + { + "name": "terminal-bench/build-pov-ray", + "ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb" + }, + { + "name": "terminal-bench/caffe-cifar-10", + "ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf" + }, + { + "name": "terminal-bench/cancel-async-tasks", + "ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1" + }, + { + "name": "terminal-bench/circuit-fibsqrt", + "ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3" + }, + { + "name": "terminal-bench/cobol-modernization", + "ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88" + }, + { + "name": "terminal-bench/constraints-scheduling", + "ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53" + }, + { + "name": "terminal-bench/custom-memory-heap-crash", + "ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e" + }, + { + "name": "terminal-bench/db-wal-recovery", + "ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4" + }, + { + "name": "terminal-bench/dna-assembly", + "ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f" + }, + { + "name": "terminal-bench/fix-code-vulnerability", + "ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71" + }, + { + "name": "terminal-bench/git-multibranch", + "ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa" + }, + { + "name": "terminal-bench/kv-store-grpc", + "ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307" + }, + { + "name": "terminal-bench/llm-inference-batching-scheduler", + "ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67" + }, + { + "name": "terminal-bench/log-summary-date-ranges", + "ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6" + }, + { + "name": "terminal-bench/merge-diff-arc-agi-task", + "ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d" + }, + { + "name": "terminal-bench/model-extraction-relu-logits", + "ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17" + }, + { + "name": "terminal-bench/mteb-leaderboard", + "ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c" + }, + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + }, + { + "name": "terminal-bench/nginx-request-logging", + "ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097" + }, + { + "name": "terminal-bench/openssl-selfsigned-cert", + "ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302" + }, + { + "name": "terminal-bench/path-tracing", + "ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c" + }, + { + "name": "terminal-bench/pypi-server", + "ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1" + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + }, + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + }, + { + "name": "terminal-bench/qemu-alpine-ssh", + "ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28" + }, + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + }, + { + "name": "terminal-bench/regex-chess", + "ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78" + }, + { + "name": "terminal-bench/regex-log", + "ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a" + }, + { + "name": "terminal-bench/schemelike-metacircular-eval", + "ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6" + }, + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + }, + { + "name": "terminal-bench/write-compressor", + "ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e" + } + ], + "prior_exposure": "All tasks are development tasks with previous outcomes and some reviewed traces/tests. This is not a held-out capability evaluation. No fresh trace or hidden-test content is reviewed until every new agent output is frozen.", + "remaining_unopened_tasks": 55 +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json new file mode 100644 index 0000000..313bbe0 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json @@ -0,0 +1,29 @@ +{ + "attempts": [ + { + "job_name": "dsh-dev34-smoke-install-20261030", + "stage": "agent_install", + "status": "setup_failed", + "model_called": false, + "cause": "The Python runtime dsh command requires an explicit DSH_HOME even for --version.", + "correction": "Set DSH_HOME=/tmp/harbor-dsh-version on the version probe. SDK task homes remain independently isolated.", + "evidence_sha256": { + "jobs/dsh-dev34-smoke-install-20261030/db-wal-recovery__yp998NF/exception.txt": "6b7401ada06fbd91734f7e82267e10c6677ac36ca72095553d86217e5e534934", + "jobs/dsh-dev34-smoke-install-20261030/db-wal-recovery__yp998NF/agent/bootstrap.json": "8ba79d5fac285504340a8b6cc35e43353935858a3b52b971935e6eddc6cd41ce", + "jobs/dsh-dev34-smoke-install-20261030-input/summary.json": "8a46fef3af0e76b7e236408d99c0a48a7c1d0c6be1b9118a6245803a878ea658" + } + }, + { + "job_name": "dsh-dev34-smoke-install-v2-20261030", + "status": "ready", + "model_called": false, + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "evidence_sha256": { + "jobs/dsh-dev34-smoke-install-v2-20261030/db-wal-recovery__NoW52ML/agent/bootstrap.json": "317eb4d1a905a9446f48890f4be51b9acf3c0e230c0770b360285bae70d579b1", + "jobs/dsh-dev34-smoke-install-v2-20261030/db-wal-recovery__NoW52ML/result.json": "c8a5d4bdcdfa2496624f04041121fa0ddbc6ea6d1a5a460bf7da387ec0cb5a08", + "jobs/dsh-dev34-smoke-install-v2-20261030-input/summary.json": "8b1f7cd1227b87075a7f1a182c3d8d3ba79ed88ccef205d8c84081d2acd36f94" + } + } + ] +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/smoke-install.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/smoke-install.json new file mode 100644 index 0000000..72a65d2 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/smoke-install.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/db-wal-recovery", + "ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-pipeline-parallelism.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-pipeline-parallelism.json new file mode 100644 index 0000000..f4df99b --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-pipeline-parallelism.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-tensor-parallelism.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-tensor-parallelism.json new file mode 100644 index 0000000..f3cc92a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/torch-tensor-parallelism.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/write-compressor.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/write-compressor.json new file mode 100644 index 0000000..c0e505c --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/write-compressor.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/write-compressor", + "ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} From a227c5cdb1492f1fb4fa317381beb5501a6ee4e5 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Fri, 30 Oct 2026 09:03:45 -0700 Subject: [PATCH 04/15] docs(benchmarks): correct the registered verifier profile metadata --- .../deepseek-harness-dev34-20261030/README.md | 16 ++++++++++-- .../profile-metadata-amendment.json | 25 +++++++++++++++++++ 2 files changed, 39 insertions(+), 2 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md index a0c3188..61ffa9e 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md @@ -37,8 +37,8 @@ reminders. All tasks require fresh installation-only checks before the model queue opens. These checks install hash-pinned SDK/runtime wheels and their Python dependencies, initialize the actual SDK profile/provider/model without a task -prompt, and exercise the Linux process supervisor. QEMU additionally runs its -reviewed verifier dependency profile. The other 33 verifier profiles are +prompt, and exercise the Linux process supervisor. QEMU and PyTorch model +recovery additionally run their reviewed verifier dependency profiles. The other 32 verifier profiles are explicitly `not_configured`; agent installation success does not certify them. The existing APT-index and original uv/Python release caches retain upstream @@ -109,3 +109,15 @@ from the frozen source root. Native outputs and per-job input manifests remain under `jobs/deepseek-harness-runtime-source/jobs/`. Orchestration and image-cache events remain under `jobs/deepseek-harness-dev34-20261030-orchestration/`. The runner refuses to overwrite or replay previously started model jobs. + +## Verifier profile metadata correction + +The [metadata amendment](profile-metadata-amendment.json), registered before +model work, corrects a historical Pi description copied into `experiment.json`. +The frozen current adapter registers both `qemu-alpine-ssh-v1` and +`pytorch-model-recovery-v1` for these selected revisions, and automatically +enforces both. The other 32 tasks have no registered verifier profile. This +corrects the plan's descriptive profile fields only; it changes no execution, +task, source, dependency, timeout, or model setting. The original plan and its +execution registration remain immutable. QEMU's profile has passed; PyTorch's +profile is running. Native bootstrap records determine the final status. diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json new file mode 100644 index 0000000..72030d8 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json @@ -0,0 +1,25 @@ +{ + "schema_version": 1, + "registered_at": "2026-10-30T16:02:46.103793+00:00", + "original_plan": "experiment.json", + "original_plan_sha256": "fb59d5618c44e691babb80b2ea8323eb45edf9f45383e61344d747b870bc6d6e", + "reason": "The copied historical Pi plan described only the QEMU verifier profile. The frozen current adapter also registers the selected pytorch-model-recovery revision. Both profiles have been automatically enforced by the actual run.", + "changes_execution": false, + "model_trials_started": 0, + "authoritative_profile_metadata": { + "terminal-bench/qemu-alpine-ssh": { + "ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28", + "profile": "qemu-alpine-ssh-v1" + }, + "terminal-bench/pytorch-model-recovery": { + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "profile": "pytorch-model-recovery-v1" + } + }, + "other_task_count": 32, + "other_task_profile_status": "not_configured", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "source_file": "benchmarks/harbor/src/harbor_adapter/bootstrap.py", + "source_sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "interpretation": "This supersedes only the original plan preflight_policy text and its pytorch-model-recovery verifier_dependency_profile metadata. Task revisions, resources, source, images, model settings, generation policy, configurations, execution order, and retry policy are unchanged. Keep the original registration immutable. Actual success must be established from each native bootstrap record." +} From f88b3b621430216e1fd243f122d15688817c48c7 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Fri, 30 Oct 2026 09:21:44 -0700 Subject: [PATCH 05/15] docs(benchmarks): record successful deepseek preflight checks --- .../deepseek-harness-dev34-20261030/README.md | 12 +- .../preflight-evidence.json | 1215 +++++++++++++++++ 2 files changed, 1223 insertions(+), 4 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/preflight-evidence.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md index 61ffa9e..6d08b55 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md @@ -1,7 +1,11 @@ # DeepSeek Harness on the Pi 34-task development pool -Status: the 34-task installation phase is running; no model trials -have started. The official keyless SDK smoke and corrected WAL container +Status: all 34 installation checks passed. Model execution has not started; +using the existing Pi-configured DeepSeek credential is awaiting explicit +maintainer authorization after automatic approval review blocked that source. +The [preflight audit](preflight-evidence.json) records SDK initialization and +process supervision success for every task, two passed verifier dependency +profiles, and 32 explicitly unconfigured profiles. The official keyless SDK smoke and corrected WAL container installation smoke passed. The first WAL setup failure is preserved in [setup evidence](setup-evidence.json). @@ -119,5 +123,5 @@ The frozen current adapter registers both `qemu-alpine-ssh-v1` and enforces both. The other 32 tasks have no registered verifier profile. This corrects the plan's descriptive profile fields only; it changes no execution, task, source, dependency, timeout, or model setting. The original plan and its -execution registration remain immutable. QEMU's profile has passed; PyTorch's -profile is running. Native bootstrap records determine the final status. +execution registration remain immutable. Both QEMU's and PyTorch's dependency +profiles passed their native preflight checks. diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/preflight-evidence.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/preflight-evidence.json new file mode 100644 index 0000000..3fdbbae --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/preflight-evidence.json @@ -0,0 +1,1215 @@ +{ + "schema_version": 1, + "audited_at": "2026-10-30T16:20:53.637565+00:00", + "model_jobs_started": 0, + "experiment": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json", + "sha256": "fb59d5618c44e691babb80b2ea8323eb45edf9f45383e61344d747b870bc6d6e" + }, + "profile_metadata_amendment": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json", + "sha256": "f27cf8b8939506f1ec68d964322aa9ef4af6da4a0f2e9f95c1b32aaed59fb7de" + }, + "adapter_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "summary": { + "installation_passed": 34, + "sdk_initialization_passed": 34, + "supervision_probe_passed": 34, + "verifier_dependency_profiles_passed": 2, + "verifier_dependency_profiles_not_configured": 32, + "model_executions": 0 + }, + "trials": [ + { + "task": "terminal-bench/adaptive-rejection-sampler", + "task_ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de", + "job": "dsh-dev34-install-pair-01-20261030", + "trial": "adaptive-rejection-sampler__BuqYqRV", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-01-20261030.json", + "sha256": "35b97de60af32fd25dcb0223dfc5cade5f491cc0361a2424e7973cde21826860" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/manifest.json", + "sha256": "a47999c0d83ce04d3b6ace156f7e73b9763cd4e495e9c51708547b99285a07ca" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/summary.json", + "sha256": "61bae69a5c87321a32a782a96fcd43e627d9c45049a7cf9cb24648e5d1143a31" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/result.json", + "sha256": "9543b465485a93422a68d7741ee610603545d2e9b41cec779106bdef39c4d225" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/build-cython-ext", + "task_ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12", + "job": "dsh-dev34-install-pair-01-20261030", + "trial": "build-cython-ext__4W44WcX", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-01-20261030.json", + "sha256": "35b97de60af32fd25dcb0223dfc5cade5f491cc0361a2424e7973cde21826860" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/manifest.json", + "sha256": "a47999c0d83ce04d3b6ace156f7e73b9763cd4e495e9c51708547b99285a07ca" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/summary.json", + "sha256": "61bae69a5c87321a32a782a96fcd43e627d9c45049a7cf9cb24648e5d1143a31" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/result.json", + "sha256": "cc060ca2421d5d182bc554c9f5f9f003521ec0f4a25b9fc3fbce0bc6e0edc8f7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/agent/bootstrap.json", + "sha256": "0c82eb159d150060290eebcfbe93607927cd4537bc16959451b2dcc648d61767" + } + ] + }, + { + "task": "terminal-bench/cancel-async-tasks", + "task_ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1", + "job": "dsh-dev34-install-pair-02-20261030", + "trial": "cancel-async-tasks__cagFx9t", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-02-20261030.json", + "sha256": "b5a81d0535ecaaac098b9b5b4839c5d5946950d70de5aba646fd27bd700320d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/manifest.json", + "sha256": "0a7a13cc673070c646be8d06fd8fbd2013c06f32a1fff6131d03d1c77755f95c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/summary.json", + "sha256": "1d57bea97ec7cd5d3cc348e065195ac5579e7d1a3786a177b8b3c5a0e2389a42" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/result.json", + "sha256": "880f693138aae3abba7bfdb0a178aaadbc8c9629e329656d4ccda8ee84a93d1e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/cobol-modernization", + "task_ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88", + "job": "dsh-dev34-install-pair-02-20261030", + "trial": "cobol-modernization__Uorq3pW", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-02-20261030.json", + "sha256": "b5a81d0535ecaaac098b9b5b4839c5d5946950d70de5aba646fd27bd700320d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/manifest.json", + "sha256": "0a7a13cc673070c646be8d06fd8fbd2013c06f32a1fff6131d03d1c77755f95c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/summary.json", + "sha256": "1d57bea97ec7cd5d3cc348e065195ac5579e7d1a3786a177b8b3c5a0e2389a42" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/result.json", + "sha256": "92b26a3a93ff64857ecd4b6d65b6853308aefb3eb3f190614967d1a790f3a567" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/agent/bootstrap.json", + "sha256": "d75e36e5346e74a3f1b464b24b6e3bbf159103f626e474d7833ffed5bb202221" + } + ] + }, + { + "task": "terminal-bench/db-wal-recovery", + "task_ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4", + "job": "dsh-dev34-install-pair-03-20261030", + "trial": "db-wal-recovery__SsWt56x", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-03-20261030.json", + "sha256": "7997484bc8cf759eac5ad91dcbc9aee1a310d1ab9de5d962aa2dab6f4141e2ad" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/manifest.json", + "sha256": "2f35cbc799352d952caabace8f9ed3eb9f9d6bea62f390088e9c92698362d61e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/summary.json", + "sha256": "90943816f7581a3676aa4a7c4f6052ec055de61c1de145e8c58c17af03fc400f" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/result.json", + "sha256": "625c9a6e005e000984e7fc8111f481fd86ee259893d3c8f5cffe18b4b6291160" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/agent/bootstrap.json", + "sha256": "317eb4d1a905a9446f48890f4be51b9acf3c0e230c0770b360285bae70d579b1" + } + ] + }, + { + "task": "terminal-bench/fix-code-vulnerability", + "task_ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71", + "job": "dsh-dev34-install-pair-03-20261030", + "trial": "fix-code-vulnerability__7tYLPYo", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-03-20261030.json", + "sha256": "7997484bc8cf759eac5ad91dcbc9aee1a310d1ab9de5d962aa2dab6f4141e2ad" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/manifest.json", + "sha256": "2f35cbc799352d952caabace8f9ed3eb9f9d6bea62f390088e9c92698362d61e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/summary.json", + "sha256": "90943816f7581a3676aa4a7c4f6052ec055de61c1de145e8c58c17af03fc400f" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/result.json", + "sha256": "26f8e8a79707a4f334422558031a0e2faafef53aad085ca04a688f8a2623a926" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/agent/bootstrap.json", + "sha256": "a243941e9357e36a407ff4cfd4cf32498b67756cf083283184e3ac20269ed125" + } + ] + }, + { + "task": "terminal-bench/git-multibranch", + "task_ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa", + "job": "dsh-dev34-install-pair-04-20261030", + "trial": "git-multibranch__uksBvS7", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-04-20261030.json", + "sha256": "08cc0bb602fdc5741e4432833ada51081e2d311d3030c63b700961c658545130" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/manifest.json", + "sha256": "e46c64ccb95f0ed5b47e11db1c4404df71ab21a2096ddfc3df560a3e3c8a36d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/summary.json", + "sha256": "1262258e1c174e5c3e0cae701841e515c4e61714297ad141f6ba5d7ce637ce4a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/result.json", + "sha256": "aa8f47a7aa4d3e558ef34fb2ee22f9ec7cdcf7c44f4eb0a54c64e5d204384ac5" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/agent/bootstrap.json", + "sha256": "139a6fd0776dee86f5a0306ebdadb3f1a4e79938a27aa2a5edcdd96dfa9be907" + } + ] + }, + { + "task": "terminal-bench/kv-store-grpc", + "task_ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307", + "job": "dsh-dev34-install-pair-04-20261030", + "trial": "kv-store-grpc__V6nNm5R", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-04-20261030.json", + "sha256": "08cc0bb602fdc5741e4432833ada51081e2d311d3030c63b700961c658545130" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/manifest.json", + "sha256": "e46c64ccb95f0ed5b47e11db1c4404df71ab21a2096ddfc3df560a3e3c8a36d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/summary.json", + "sha256": "1262258e1c174e5c3e0cae701841e515c4e61714297ad141f6ba5d7ce637ce4a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/result.json", + "sha256": "ed6e15f3638c215e50b21a7dce5a86eebcbbb07ad5778fa02eace1d89371baad" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/log-summary-date-ranges", + "task_ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6", + "job": "dsh-dev34-install-pair-05-20261030", + "trial": "log-summary-date-ranges__6dcMfYE", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-05-20261030.json", + "sha256": "b75fd05b46d16329753c582920cd1b08b75695f6a4728f293ecbc6bb2dcd20ed" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/manifest.json", + "sha256": "09adca648d3cd50c864e95c0b3d36326b7db09b5227435580427cccd893102bd" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/summary.json", + "sha256": "0196f9f1298444e0a42ff5070ccd7f89717ba14cb8bc5edb34ec404e066de7b4" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/result.json", + "sha256": "2c5d6232aeec9ee5b4c040a0a9ef0e5e76f360bf880579cba4bae9d10b69ad5d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/merge-diff-arc-agi-task", + "task_ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d", + "job": "dsh-dev34-install-pair-05-20261030", + "trial": "merge-diff-arc-agi-task__RGHcv6Z", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-05-20261030.json", + "sha256": "b75fd05b46d16329753c582920cd1b08b75695f6a4728f293ecbc6bb2dcd20ed" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/manifest.json", + "sha256": "09adca648d3cd50c864e95c0b3d36326b7db09b5227435580427cccd893102bd" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/summary.json", + "sha256": "0196f9f1298444e0a42ff5070ccd7f89717ba14cb8bc5edb34ec404e066de7b4" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/result.json", + "sha256": "2cdf1b3a5485cdbc6db2f390c5186dec2939d374c476543f04edad59a0c66a14" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/model-extraction-relu-logits", + "task_ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17", + "job": "dsh-dev34-install-pair-06-20261030", + "trial": "model-extraction-relu-logits__Ky6SN6a", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-06-20261030.json", + "sha256": "f7c8461d95344962910447b29e0b4b942794d2aeb5d854e245100ba2f7b5a45b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/manifest.json", + "sha256": "c0f6916a703f0c9110e958bf0f797a2b4a261f65c6be3fc137d3a54daf01db73" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/summary.json", + "sha256": "b13d0543f0cb1c0f367e6f74317c2fe759e399297e2061665719c64847dfdb22" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/result.json", + "sha256": "1cf09cef6d5f195c6f64e1dddf49fe52fe61583e8a6ffaf75bb9e6454ae2c99e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/multi-source-data-merger", + "task_ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "job": "dsh-dev34-install-pair-06-20261030", + "trial": "multi-source-data-merger__pxBy43s", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-06-20261030.json", + "sha256": "f7c8461d95344962910447b29e0b4b942794d2aeb5d854e245100ba2f7b5a45b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/manifest.json", + "sha256": "c0f6916a703f0c9110e958bf0f797a2b4a261f65c6be3fc137d3a54daf01db73" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/summary.json", + "sha256": "b13d0543f0cb1c0f367e6f74317c2fe759e399297e2061665719c64847dfdb22" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/result.json", + "sha256": "cad316ea002c327d02d2eebdcb379f5cc2b631396cefc82cb34d306f4e92c16c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/agent/bootstrap.json", + "sha256": "38a018af813827b21bff5685e056300a44dc7899ae803e02ae564568808be3fd" + } + ] + }, + { + "task": "terminal-bench/nginx-request-logging", + "task_ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097", + "job": "dsh-dev34-install-pair-07-20261030", + "trial": "nginx-request-logging__ra8o6iV", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-07-20261030.json", + "sha256": "8dc0becd58a3de76db3fbb638e01e6a0296fd3680cc37c50cec9288f94d57d41" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/manifest.json", + "sha256": "59297ca64e29863872cfa97ab882a77328d38ad7f5584a6890489c8825f81800" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/summary.json", + "sha256": "37b07dd932adf6c214a49be3fcc88a67ee069f1d178d036b372829d6ffe85a48" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/result.json", + "sha256": "5f06c592c06135e40b91172af316333c4edca880235a1f386784ea97acc24d59" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/agent/bootstrap.json", + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e" + } + ] + }, + { + "task": "terminal-bench/openssl-selfsigned-cert", + "task_ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302", + "job": "dsh-dev34-install-pair-07-20261030", + "trial": "openssl-selfsigned-cert__6stQ3ja", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-07-20261030.json", + "sha256": "8dc0becd58a3de76db3fbb638e01e6a0296fd3680cc37c50cec9288f94d57d41" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/manifest.json", + "sha256": "59297ca64e29863872cfa97ab882a77328d38ad7f5584a6890489c8825f81800" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/summary.json", + "sha256": "37b07dd932adf6c214a49be3fcc88a67ee069f1d178d036b372829d6ffe85a48" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/result.json", + "sha256": "6b07e89c5e841b478bfc4241ecf3808dd391f9cff3e3e95dff16d50a6d28823b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/pypi-server", + "task_ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1", + "job": "dsh-dev34-install-pair-08-20261030", + "trial": "pypi-server__4jBWBVR", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-08-20261030.json", + "sha256": "9d4230aae5324a094c5d2ec43236b6cb651cd777805d46821527cdd8c607d4eb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/manifest.json", + "sha256": "97bdcff4c34faf3d787c3860d68f8b958812d6ae51f20ed00012b62abb4da721" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/summary.json", + "sha256": "53288df2ea90779162c20bd6ebdfc7b38a8d38f90acf3794a9e47f2ad711e2a7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/result.json", + "sha256": "73709f5a967b2caf1082b5a45144fa483bfdb0643f599e792db0ad5da11f3a9c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/agent/bootstrap.json", + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-cli", + "task_ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "job": "dsh-dev34-install-pair-08-20261030", + "trial": "pytorch-model-cli__Lnfimqu", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-08-20261030.json", + "sha256": "9d4230aae5324a094c5d2ec43236b6cb651cd777805d46821527cdd8c607d4eb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/manifest.json", + "sha256": "97bdcff4c34faf3d787c3860d68f8b958812d6ae51f20ed00012b62abb4da721" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/summary.json", + "sha256": "53288df2ea90779162c20bd6ebdfc7b38a8d38f90acf3794a9e47f2ad711e2a7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/result.json", + "sha256": "a50ce1fd10a899155353e312aa5ef356fb05194cfba6f73e899c9e9da4fec620" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "task_ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "job": "dsh-dev34-install-pair-09-20261030", + "trial": "pytorch-model-recovery__FCnSvrf", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "passed" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-09-20261030.json", + "sha256": "cee92389b124bfbca5c1d05a1c3cd1467b2986ef9a6bef7248537dcc3ba81f93" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/manifest.json", + "sha256": "00a879f86f550d9aea4530e2fd058f04890540a26c0b2a15f44b2d5f50b01803" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/summary.json", + "sha256": "4f447719c1b8d78ab9aede92dddd9145abfbfb292efc75b40e7cffac7184ec45" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/result.json", + "sha256": "6a7bf4ba88b91f90a6cd845403006de53b7ac3ad2df1db84d8f09ca0125730ef" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/agent/bootstrap.json", + "sha256": "b97b124b61cfc3b413b29b69460372d0b964474f0b96133c4a5571d9df639c45" + } + ] + }, + { + "task": "terminal-bench/qemu-alpine-ssh", + "task_ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28", + "job": "dsh-dev34-install-pair-09-20261030", + "trial": "qemu-alpine-ssh__eb5Y6dk", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "qemu-alpine-ssh-v1", + "status": "passed" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-09-20261030.json", + "sha256": "cee92389b124bfbca5c1d05a1c3cd1467b2986ef9a6bef7248537dcc3ba81f93" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/manifest.json", + "sha256": "00a879f86f550d9aea4530e2fd058f04890540a26c0b2a15f44b2d5f50b01803" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/summary.json", + "sha256": "4f447719c1b8d78ab9aede92dddd9145abfbfb292efc75b40e7cffac7184ec45" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/result.json", + "sha256": "0e3ccd245745a31c01eadf8d10e4ae5a061d083fc07e828211a99560cf9196c3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/agent/bootstrap.json", + "sha256": "1e672441c98d4e4e39bbb460a3c2c7e1d185954628be757adf5af4f0eef9a767" + } + ] + }, + { + "task": "terminal-bench/query-optimize", + "task_ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "job": "dsh-dev34-install-pair-10-20261030", + "trial": "query-optimize__AbiSxXy", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-10-20261030.json", + "sha256": "b70a073249ebb1d2a688c3a6f5f068519f67beef0423e5179f966bc93e866138" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/manifest.json", + "sha256": "502459c6482d86928087898c676b5390c2229d0009c952fb02da783352bcf51a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/summary.json", + "sha256": "d6654fb376206566eb2d98c82cbb9f0a3ecbe0047456c5e74878530ca465e258" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/result.json", + "sha256": "6fa73803640f1d1bc557ef8bac23af91149d7f55c10bb5c7112e45239115812c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/agent/bootstrap.json", + "sha256": "89006797609af114ee0903b557c5ff434b3ef6071d8346100829e653fceb38ca" + } + ] + }, + { + "task": "terminal-bench/regex-log", + "task_ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a", + "job": "dsh-dev34-install-pair-10-20261030", + "trial": "regex-log__VD4fbS3", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-10-20261030.json", + "sha256": "b70a073249ebb1d2a688c3a6f5f068519f67beef0423e5179f966bc93e866138" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/manifest.json", + "sha256": "502459c6482d86928087898c676b5390c2229d0009c952fb02da783352bcf51a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/summary.json", + "sha256": "d6654fb376206566eb2d98c82cbb9f0a3ecbe0047456c5e74878530ca465e258" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/result.json", + "sha256": "091663f68b91509c2337d427b83e75b3b4a7df9bfcad6e94afd4ec565831af3f" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "task_ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "job": "dsh-dev34-install-pair-11-20261030", + "trial": "torch-pipeline-parallelism__AmhPcjR", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-11-20261030.json", + "sha256": "0b35669d6acce8ac74eb247f50d00c36a944b274020be486bfd49821eb7abcc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/manifest.json", + "sha256": "8b148e8e8bc4495d1806a39c7e4e37d1690172ff925179890fd0918f4583365d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/summary.json", + "sha256": "2014331be4e6536020bbba6ae8705958c391f21cf40b4cbf45d042eebad3a727" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/result.json", + "sha256": "bc184f6f88666ccdfe68e25811e45b5ebcacc321b247307bb88bff5408efd968" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "task_ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "job": "dsh-dev34-install-pair-11-20261030", + "trial": "torch-tensor-parallelism__NVe5csr", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-11-20261030.json", + "sha256": "0b35669d6acce8ac74eb247f50d00c36a944b274020be486bfd49821eb7abcc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/manifest.json", + "sha256": "8b148e8e8bc4495d1806a39c7e4e37d1690172ff925179890fd0918f4583365d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/summary.json", + "sha256": "2014331be4e6536020bbba6ae8705958c391f21cf40b4cbf45d042eebad3a727" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/result.json", + "sha256": "5ee413a851303fa81a99928b7a8b2653bf933f2452672b1b52841cc827edaf7c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/constraints-scheduling", + "task_ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53", + "job": "dsh-dev34-install-pair-12-20261030", + "trial": "constraints-scheduling__4EcXjX2", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-12-20261030.json", + "sha256": "bef506c7122cdd23faccdea2f9631b32c7dbc89fbdf961313f822c9aaee85c7a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/manifest.json", + "sha256": "93bcf89c9a1af9248e825cca85eb2bade9579b60a8a79fb12ff0ce7e8870537d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/summary.json", + "sha256": "6f527aa51e0909c36993a35e0784ed31c01418d97d8355d745e979cda8351577" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/result.json", + "sha256": "6aacea30f7ff30f3913b0b76892263bf23add9253d86e0f76ea0b8fe4b1067fd" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/agent/bootstrap.json", + "sha256": "cf2881c5ea9a508fcf319b2a6befc60d39bd25004a7d4f3d2b194579b31fe401" + } + ] + }, + { + "task": "terminal-bench/write-compressor", + "task_ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e", + "job": "dsh-dev34-install-pair-12-20261030", + "trial": "write-compressor__HFAgpjW", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-12-20261030.json", + "sha256": "bef506c7122cdd23faccdea2f9631b32c7dbc89fbdf961313f822c9aaee85c7a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/manifest.json", + "sha256": "93bcf89c9a1af9248e825cca85eb2bade9579b60a8a79fb12ff0ce7e8870537d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/summary.json", + "sha256": "6f527aa51e0909c36993a35e0784ed31c01418d97d8355d745e979cda8351577" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/result.json", + "sha256": "1e5a5194aedd3712d9af22c4a6c25124068a632a8e8846e7c7700820acada7a1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/agent/bootstrap.json", + "sha256": "106fca350e150fcc87b6003fc699eec6e057797103b83a4c169cf5c14a6b293c" + } + ] + }, + { + "task": "terminal-bench/custom-memory-heap-crash", + "task_ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e", + "job": "dsh-dev34-install-pair-13-20261030", + "trial": "custom-memory-heap-crash__DQuhNvN", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-13-20261030.json", + "sha256": "88e9bcb621608f17e1c371707074425eab1e5e60a56c7999eced6cbdf394cb65" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/manifest.json", + "sha256": "eb140223825530b724c6b4a7455a51a44b4a6d9a0c0048a20bab069bf7111446" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/summary.json", + "sha256": "995297252ae987cf60c3d4b780ab768e4b51f1bf0bdb0c2327fddcf92ba9547c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/result.json", + "sha256": "8e1e193dd8d4b1bfeb8cb65a45a488e946deabf848ea8322e28bc630c0465449" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/agent/bootstrap.json", + "sha256": "747599891fefaac25e7ab1119fee62bb88a9bf9b3e036a496c37c79d15610b74" + } + ] + }, + { + "task": "terminal-bench/dna-assembly", + "task_ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f", + "job": "dsh-dev34-install-pair-13-20261030", + "trial": "dna-assembly__38WuERG", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-13-20261030.json", + "sha256": "88e9bcb621608f17e1c371707074425eab1e5e60a56c7999eced6cbdf394cb65" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/manifest.json", + "sha256": "eb140223825530b724c6b4a7455a51a44b4a6d9a0c0048a20bab069bf7111446" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/summary.json", + "sha256": "995297252ae987cf60c3d4b780ab768e4b51f1bf0bdb0c2327fddcf92ba9547c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/result.json", + "sha256": "b005dd171d4df9c4d790f4518fad8122370217d3252fa14b9bd6cd4c21508aa7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/llm-inference-batching-scheduler", + "task_ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67", + "job": "dsh-dev34-install-pair-14-20261030", + "trial": "llm-inference-batching-scheduler__kd5nP2N", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-14-20261030.json", + "sha256": "e6b14ee3f5a53b3bd0d4b6c54b92db8fc78046dfe82ed834985bccc8bdb56c64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/manifest.json", + "sha256": "86d588aafeabba43bd931a297bc78b960761b2f9b878bece05301deab3defad3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/summary.json", + "sha256": "2cef49e9735b839280086ea02251978e776fbebbde556d0bb5ecc2dcbebbaa64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/result.json", + "sha256": "d9d86f9778f07453b8949b1eafe18f8349d059f247d623410e90f63c9a504561" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/path-tracing", + "task_ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c", + "job": "dsh-dev34-install-pair-14-20261030", + "trial": "path-tracing__2oz9xrm", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-14-20261030.json", + "sha256": "e6b14ee3f5a53b3bd0d4b6c54b92db8fc78046dfe82ed834985bccc8bdb56c64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/manifest.json", + "sha256": "86d588aafeabba43bd931a297bc78b960761b2f9b878bece05301deab3defad3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/summary.json", + "sha256": "2cef49e9735b839280086ea02251978e776fbebbde556d0bb5ecc2dcbebbaa64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/result.json", + "sha256": "0878bb05a625a7f36d0f90fd39f02e30e522daa6af9781399da7f34930b0aedb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/agent/bootstrap.json", + "sha256": "8a3916db45911667d182143670fe4edfb134d999147e568a8af01013ff22ca93" + } + ] + }, + { + "task": "terminal-bench/caffe-cifar-10", + "task_ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf", + "job": "dsh-dev34-install-pair-15-20261030", + "trial": "caffe-cifar-10__gofkboQ", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-15-20261030.json", + "sha256": "7d3111e619275c66f4adad6cb420171e1d1c5ae80401b38288dc830874f313de" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/manifest.json", + "sha256": "2c81a93d2ba080cb4bcc3c359eb10e95b4074eb5c324f2076fb5db330fab8942" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/summary.json", + "sha256": "be4267f0ea72f63625f79ab7f8c43a598fb2eb358464564bda289a9b8e3db02b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/result.json", + "sha256": "56e49dc1210e839e68a550346dd1011637a9780345c7a5a9cdf84fea3c360e75" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/agent/bootstrap.json", + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d" + } + ] + }, + { + "task": "terminal-bench/schemelike-metacircular-eval", + "task_ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6", + "job": "dsh-dev34-install-pair-15-20261030", + "trial": "schemelike-metacircular-eval__sesEXMa", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-15-20261030.json", + "sha256": "7d3111e619275c66f4adad6cb420171e1d1c5ae80401b38288dc830874f313de" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/manifest.json", + "sha256": "2c81a93d2ba080cb4bcc3c359eb10e95b4074eb5c324f2076fb5db330fab8942" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/summary.json", + "sha256": "be4267f0ea72f63625f79ab7f8c43a598fb2eb358464564bda289a9b8e3db02b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/result.json", + "sha256": "a13296c3bbd5cfb2e2e6e7c0e37d360492178221e49a2cd868078abe876d7629" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/circuit-fibsqrt", + "task_ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3", + "job": "dsh-dev34-install-pair-16-20261030", + "trial": "circuit-fibsqrt__6JWTyKR", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-16-20261030.json", + "sha256": "03782151798a8c423bf3dd80c0381749060f14443fa18573990d0a52cece5bc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/manifest.json", + "sha256": "757978d6b034758a75027aac61264ae6eb72243da9338ffafdeae28cfa78051b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/summary.json", + "sha256": "40055254b4c8127df523ee89e946b0426f1f65b2935643c17bb447d5f40f3a5c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/result.json", + "sha256": "2a6b3a5ec5956d1249a5663543ac1149cb4a1fecfaadd0b7c34a6a1721939c60" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/agent/bootstrap.json", + "sha256": "933708cbc5cd3c97f336c849d1eee8af9f3b824eddd6aa7506dc53f1657d8247" + } + ] + }, + { + "task": "terminal-bench/mteb-leaderboard", + "task_ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c", + "job": "dsh-dev34-install-pair-16-20261030", + "trial": "mteb-leaderboard__KCD6tEf", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-16-20261030.json", + "sha256": "03782151798a8c423bf3dd80c0381749060f14443fa18573990d0a52cece5bc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/manifest.json", + "sha256": "757978d6b034758a75027aac61264ae6eb72243da9338ffafdeae28cfa78051b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/summary.json", + "sha256": "40055254b4c8127df523ee89e946b0426f1f65b2935643c17bb447d5f40f3a5c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/result.json", + "sha256": "06d54613223f16144e4d17d9c12e0a9061487bc4fffba4d0d146ef0e55cefdfb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/agent/bootstrap.json", + "sha256": "8a30f5515bd2ccd7749dc2993f8a83b0b668c109ec78febbf08cad59e3b4caa1" + } + ] + }, + { + "task": "terminal-bench/build-pov-ray", + "task_ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb", + "job": "dsh-dev34-install-pair-17-20261030", + "trial": "build-pov-ray__TbS8MbA", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-17-20261030.json", + "sha256": "a0b54e31b9726f209c2972852b2ed83fc504600eace0db69d14235ef16e91e0a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/manifest.json", + "sha256": "f8e99d12bdfa6699177b64203869a4cee08d713a5349a251f445a60eb6eddd80" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/summary.json", + "sha256": "32a7d803060d971c6db5bfe036524455d3000efa4997f37f4e8887acc1d62e32" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/result.json", + "sha256": "f9a470171cc88e6a3728a04b99d9ef0d9abf92d9768cc9e63cbdd93686b1582c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/agent/bootstrap.json", + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d" + } + ] + }, + { + "task": "terminal-bench/regex-chess", + "task_ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78", + "job": "dsh-dev34-install-pair-17-20261030", + "trial": "regex-chess__3FtpzJU", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-17-20261030.json", + "sha256": "a0b54e31b9726f209c2972852b2ed83fc504600eace0db69d14235ef16e91e0a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/manifest.json", + "sha256": "f8e99d12bdfa6699177b64203869a4cee08d713a5349a251f445a60eb6eddd80" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/summary.json", + "sha256": "32a7d803060d971c6db5bfe036524455d3000efa4997f37f4e8887acc1d62e32" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/result.json", + "sha256": "a0d6f78c049b56f4ca2e954e982800a488134b1c3b3af8b1fbe10abc701e386d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/agent/bootstrap.json", + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86" + } + ] + } + ] +} From 30e1d614906423f0d4059b3d62e33ac6ce998c3d Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sat, 31 Oct 2026 04:03:39 -0700 Subject: [PATCH 06/15] docs(benchmarks): record the authorized deepseek model launch --- .../configs/deepseek-harness-dev34-20261030/README.md | 7 ++++--- .../deepseek-harness-dev34-20261030/launch-evidence.json | 9 +++++++++ 2 files changed, 13 insertions(+), 3 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/launch-evidence.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md index 6d08b55..d37e60c 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md @@ -1,8 +1,9 @@ # DeepSeek Harness on the Pi 34-task development pool -Status: all 34 installation checks passed. Model execution has not started; -using the existing Pi-configured DeepSeek credential is awaiting explicit -maintainer authorization after automatic approval review blocked that source. +Status: all 34 installation checks passed. The maintainer authorized use of +the existing Pi-configured DeepSeek credential on October 31, 2026, and the +registered model phase has started. Results remain pending until all outputs +are frozen and audited. The [preflight audit](preflight-evidence.json) records SDK initialization and process supervision success for every task, two passed verifier dependency profiles, and 32 explicitly unconfigured profiles. The official keyless SDK smoke and corrected WAL container diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/launch-evidence.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/launch-evidence.json new file mode 100644 index 0000000..f41aab1 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/launch-evidence.json @@ -0,0 +1,9 @@ +{ + "recorded_at": "2026-10-31T11:03:33.005233+00:00", + "authorization": "The maintainer explicitly confirmed use of the existing Pi-configured official DeepSeek API credential and API charges in the conversation on 2026-10-31.", + "first_model_job_started_at": "2026-10-31T11:03:13.156046+00:00", + "credential_handling": "Loaded only into the controller environment; no secret value is recorded in configuration or published evidence. Raw agent artifacts remain local and require post-run exposure audit.", + "plan_unchanged": true, + "preflight_evidence": "preflight-evidence.json", + "model_phase_log": "jobs/deepseek-harness-dev34-model.log" +} From 2094733910581e44e70e348b414c6b541f169dd3 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sat, 31 Oct 2026 09:02:21 -0700 Subject: [PATCH 07/15] test(benchmarks): audit the official deepseek harness dev34 results --- benchmarks/harbor/README.md | 9 + benchmarks/harbor/README.zh-CN.md | 8 + .../deepseek-harness-dev34-20261030/README.md | 25 +- .../model-alias-check.json | 9 + .../outputs-freeze.json | 10034 ++++++++++++++++ .../sdk-lifecycle-probe.json | 18 + .../en/deepseek-harness-dev34-20261030.md | 281 + .../zh-CN/deepseek-harness-dev34-20261030.md | 252 + .../deepseek-harness-dev34-20261030.json | 5878 +++++++++ .../scripts/probe_deepseek_lifecycle.py | 80 + 10 files changed, 16589 insertions(+), 5 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-alias-check.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json create mode 100644 benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md create mode 100644 benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md create mode 100644 benchmarks/harbor/results/deepseek-harness-dev34-20261030.json create mode 100644 benchmarks/harbor/scripts/probe_deepseek_lifecycle.py diff --git a/benchmarks/harbor/README.md b/benchmarks/harbor/README.md index 080df0d..915d1c1 100644 --- a/benchmarks/harbor/README.md +++ b/benchmarks/harbor/README.md @@ -175,6 +175,15 @@ Reported tokens include native assistant and compaction usage; missing records and interrupted runs remain partial, and cost remains unknown. This baseline does not impose Pi's generation settings or Nano's turn and work budgets. +The first [official Harness dev34 report](reports/en/deepseek-harness-dev34-20261030.md) +records 22 passes, six scored failures, and six unscored errors on Pi's existing +34-task pool. Normal SDK closure in the pinned upstream runtime stops managed +background services, including the probe's `nohup` server. This affects service +tasks: KV and PyPI failed official checks after successful agent self-tests. +The report retains those failures and documents the +[keyless lifecycle probe](scripts/probe_deepseek_lifecycle.py). Treat these +results as measurements of the registered SDK configuration. + ### Cache original task images If Docker image pulls repeatedly stall, the optional cache utility downloads the diff --git a/benchmarks/harbor/README.zh-CN.md b/benchmarks/harbor/README.zh-CN.md index 1b249df..a2d1386 100644 --- a/benchmarks/harbor/README.zh-CN.md +++ b/benchmarks/harbor/README.zh-CN.md @@ -139,6 +139,14 @@ Python 3.13 和冻结的 wheel 集合。仅安装检查会初始化真实 SDK 压缩请求用量;缺失记录和中断运行仍标记为部分统计,费用保持未知。 此 baseline 不强制套用 Pi 的生成设置或 Nano 的轮数和工作时间预算。 +首份[官方 Harness dev34 报告](reports/zh-CN/deepseek-harness-dev34-20261030.md) +在 Pi 已使用的 34 道题上记录了 22 道通过、六道计分失败、六道因错误不计分。 +固定版本的上游运行时在正常关闭 SDK 时会停止托管的后台服务,包括探针中 +通过 `nohup` 启动的服务。这会影响服务题:KV 和 PyPI 在 agent 自测成功后, +官方检查仍然失败。报告保留这些失败,并记录了 +[无密钥生命周期探针](scripts/probe_deepseek_lifecycle.py)。这些结果衡量的是 +注册的 SDK 配置。 + ### 缓存原始任务镜像 如果 Docker 拉取镜像反复停滞,可用缓存工具从同一 Docker Hub 公共仓库分段并行下载: diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md index d37e60c..af679eb 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/README.md @@ -1,9 +1,23 @@ # DeepSeek Harness on the Pi 34-task development pool -Status: all 34 installation checks passed. The maintainer authorized use of -the existing Pi-configured DeepSeek credential on October 31, 2026, and the -registered model phase has started. Results remain pending until all outputs -are frozen and audited. +Status: completed and audited on October 31, 2026. All 34 registered jobs +finished: 22 official passes, six scored failures, and six unscored errors +(two setup, three verifier dependency, one terminal API transport error). +There were 32 model executions and no whole-trial retries. On the 26 tasks +accepted in both runs, Harness passed 21 and Pi passed 22. Both lost passes +have a confirmed SDK service-lifecycle caveat. See the complete +[English report](../../reports/en/deepseek-harness-dev34-20261030.md), +[Simplified Chinese report](../../reports/zh-CN/deepseek-harness-dev34-20261030.md), +and [audited results](../../results/deepseek-harness-dev34-20261030.json). + +All 2,506 output files were frozen before reviewing new trajectories or hidden +tests; the [freeze manifest](outputs-freeze.json) preserves their hashes. +The [keyless lifecycle diagnostic](sdk-lifecycle-probe.json) was performed +after freezing and did not replay or alter any task output. Both service +failures retain their native rewards. The maintainer authorized the existing +Pi-configured DeepSeek credential on October 31; see [launch evidence](launch-evidence.json). +All 34 initial installation checks passed, but later network-dependent setup +failed in two formal jobs. The [preflight audit](preflight-evidence.json) records SDK initialization and process supervision success for every task, two passed verifier dependency profiles, and 32 explicitly unconfigured profiles. The official keyless SDK smoke and corrected WAL container @@ -91,7 +105,8 @@ now supplies an isolated version-probe home. The corrected smoke initialized the actual SDK/profile/model and supervisor successfully. Both native records remain preserved; neither is a scored task attempt. The complete Harbor suite passed 280 tests before this small integration fix, and all 11 DeepSeek-focused -tests passed after it. +tests passed after it. The final complete Harbor suite passed all 280 tests +with local mock HTTP sockets permitted. Run from this repository root: diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-alias-check.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-alias-check.json new file mode 100644 index 0000000..215baba --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-alias-check.json @@ -0,0 +1,9 @@ +{ + "checked_at": "2026-10-31T11:23:03.042011+00:00", + "source": "https://api-docs.deepseek.com/quick_start/agent_integrations/", + "alias": "deepseek-flash", + "documented_serving_model": "DeepSeek-V4.1-Flash-1017", + "official_openai_base_url": "https://api.deepseek.com", + "official_messages_base_url": "https://api.deepseek.com/anthropic", + "limitation": "This documents the provider statement at the check date, not an immutable backend checkpoint or proof of identical serving behavior across historical runs." +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json new file mode 100644 index 0000000..23f7c23 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json @@ -0,0 +1,10034 @@ +{ + "schema_version": 1, + "frozen_at": "2026-10-31T15:47:11.582274+00:00", + "registered_model_jobs": 34, + "installation_trials": 34, + "policy": "All registered jobs finished before this freeze. No new solution trajectories or hidden test contents were reviewed before the freeze. Raw logs remain local; this manifest publishes paths, sizes, and hashes only.", + "file_count": 2506, + "files": { + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json": { + "sha256": "fb59d5618c44e691babb80b2ea8323eb45edf9f45383e61344d747b870bc6d6e", + "size": 88317 + }, + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/launch-evidence.json": { + "sha256": "c520696465ca9b64ee46e2fb786945abb9043f74baf0851c828404781d2d7cf3", + "size": 650 + }, + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-alias-check.json": { + "sha256": "cb6be7e927653fa7c7bef9110262dbe9fef6ea75db65964c04a7aa3026c1c9bb", + "size": 522 + }, + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/preflight-evidence.json": { + "sha256": "06a724f349f7ddf43cff4983e203dc192a39dd08b1e21f561bfce906e90de3e9", + "size": 58061 + }, + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json": { + "sha256": "f27cf8b8939506f1ec68d964322aa9ef4af6da4a0f2e9f95c1b32aaed59fb7de", + "size": 1615 + }, + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json": { + "sha256": "118edc62f44e571ee2df5bbd41f6ab6fffc78a22908869dcaeffd11c6c9a8fbf", + "size": 1602 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-adaptive-rejection-sampler-20261030.json": { + "sha256": "07e155e9a7eb9063c06d0786611f980a43720128c06f0027c900b835d36c7d25", + "size": 29243 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-adaptive-rejection-sampler-20261030.log": { + "sha256": "70269fc10cb20c64f5ddb8efa757a9e7c090cc400242abb5fdba9b2cfa82c38b", + "size": 372 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-cython-ext-20261030.json": { + "sha256": "7399a5504eb1875249ebb046eb0008c52c5159e3a1296b1e1893a9be194cf41d", + "size": 9181 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-cython-ext-20261030.log": { + "sha256": "fa781544ef72f94a3ab22cf8686389c93b60baecc4b6360afa22826700a298f5", + "size": 291 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-pov-ray-20261030.json": { + "sha256": "d36751765bc73718db69e4aad091f44c690c38a045d437028867b434865508bf", + "size": 10504 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-pov-ray-20261030.log": { + "sha256": "325c58ef89e27ee691f53729c63047e421bd8225b9fd42af9c09426752cab395", + "size": 285 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-caffe-cifar-10-20261030.json": { + "sha256": "3e5c92478fc5b7cc715f3f9f10147d67d8b1d55c4e5210a56ad1949819c57612", + "size": 10510 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-caffe-cifar-10-20261030.log": { + "sha256": "10f8b537db873f5aae9956288f63bb37632135b9679b522934cf95a74f0e3532", + "size": 287 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cancel-async-tasks-20261030.json": { + "sha256": "0f64ab7b7f433bc1dd9a742f90871e54d7b9da64d3faeb74dcd6676a6956553c", + "size": 27183 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cancel-async-tasks-20261030.log": { + "sha256": "40ad8f602cb67a77a0fdb3cb282b6d7c51fd69de862f93399b1ff029401ce8c3", + "size": 295 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-circuit-fibsqrt-20261030.json": { + "sha256": "cf7037c85fda1367bc8fc2270f839c3b6527d7af22ca0cf906b123181d06260d", + "size": 20335 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-circuit-fibsqrt-20261030.log": { + "sha256": "c9dc0833548cc0ebd1c79a23ef791c83d391f507534727b15a5effbd7afa18f7", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cobol-modernization-20261030.json": { + "sha256": "e2eaf39089534ae6b42cfb53e2af57b0efc1e910f8ce3100cd24be07039c2f55", + "size": 20366 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cobol-modernization-20261030.log": { + "sha256": "11d9d95ca61f8a6b0cb079e863da66a7c5c2d83470610d8c7af4c44b8276cebb", + "size": 297 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-constraints-scheduling-20261030.json": { + "sha256": "f3427b83f8e867c11ec00983493c92180a837ce6746c85b454a52841c0bf83ba", + "size": 23524 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-constraints-scheduling-20261030.log": { + "sha256": "e5fb66fc2ea9f47d63c8423bd6e9d9a92b89b1d8a29801962cc41596ebc6ecb1", + "size": 303 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-custom-memory-heap-crash-20261030.json": { + "sha256": "91243e3c6af247f91c7dbfb0bb710cca80e0e0f3055c0fc509800e879236ca40", + "size": 7638 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-custom-memory-heap-crash-20261030.log": { + "sha256": "7b9fa4fd4ebbc985a512192b4f6320d89ca584c29d5a598b65ac83a2022f559a", + "size": 307 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-db-wal-recovery-20261030.json": { + "sha256": "26244bc83c6648aa9de4760f5446ec55e2491726ca8150fd1374d6159204c232", + "size": 15907 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-db-wal-recovery-20261030.log": { + "sha256": "098ef9dba2c387bfc6458a3961b4ed86f6136477501a176928767c273d548b11", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-dna-assembly-20261030.json": { + "sha256": "bce94de2fc9b3566fbf30ddcb424e5a5a95d68c124cf7eea11cafc14afe1f818", + "size": 28855 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-dna-assembly-20261030.log": { + "sha256": "651b15f126ea2e0bf73187ed79e6a96e6e5dd6e4edc9bd632b4d50320b2b69f1", + "size": 283 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-fix-code-vulnerability-20261030.json": { + "sha256": "76f1d136fddd04cf4fd5aaa913de183569eeded32018af958bd49e11de40d6bd", + "size": 5463 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-fix-code-vulnerability-20261030.log": { + "sha256": "0d1c3850fb587bbcd9681e684e66334c2beeb7f86d00e9eb519b8cd6e232743e", + "size": 303 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-git-multibranch-20261030.json": { + "sha256": "9d5aa4fcd1e0a8c742324a51a6535128cbf4a159a44451f10406f60be7b7362f", + "size": 7564 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-git-multibranch-20261030.log": { + "sha256": "c6ba9c4ec74e3e5a719a21867588a1b654162087274074a29c4bca98e33cdf9c", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-01-20261030.json": { + "sha256": "35b97de60af32fd25dcb0223dfc5cade5f491cc0361a2424e7973cde21826860", + "size": 36019 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-01-20261030.log": { + "sha256": "39673fc450de7fb981fc2d9fa11fe418d7270f7631be3b190b5d3de028e5c534", + "size": 896 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-02-20261030.json": { + "sha256": "b5a81d0535ecaaac098b9b5b4839c5d5946950d70de5aba646fd27bd700320d1", + "size": 45193 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-02-20261030.log": { + "sha256": "3df2031308cef374c368a524012c571a3314591e145fdd139a119737da05236b", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-03-20261030.json": { + "sha256": "7997484bc8cf759eac5ad91dcbc9aee1a310d1ab9de5d962aa2dab6f4141e2ad", + "size": 19273 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-03-20261030.log": { + "sha256": "d0ab53676009b627ddad09bed8b30511f334c8ad84bed29e81ce1fec0434f107", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-04-20261030.json": { + "sha256": "08cc0bb602fdc5741e4432833ada51081e2d311d3030c63b700961c658545130", + "size": 32536 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-04-20261030.log": { + "sha256": "403fbf74b273cb0c2f929f8b3216fb37fa6a4aa4ce759aff09090b091210f7db", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-05-20261030.json": { + "sha256": "b75fd05b46d16329753c582920cd1b08b75695f6a4728f293ecbc6bb2dcd20ed", + "size": 53884 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-05-20261030.log": { + "sha256": "a7a670b9c22c3210ab32d531f63cf66f48f99cc76b1d6e108cfcfda7502caa52", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-06-20261030.json": { + "sha256": "f7c8461d95344962910447b29e0b4b942794d2aeb5d854e245100ba2f7b5a45b", + "size": 49832 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-06-20261030.log": { + "sha256": "720c0bacdbae87b4933fb8e987642ae87984c21aae821638c50395fea4438ad3", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-07-20261030.json": { + "sha256": "8dc0becd58a3de76db3fbb638e01e6a0296fd3680cc37c50cec9288f94d57d41", + "size": 44975 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-07-20261030.log": { + "sha256": "4530286eee0b03d1ffc7778e969e2bb67391e69715e5a25bc2186e0a89b91b3c", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-08-20261030.json": { + "sha256": "9d4230aae5324a094c5d2ec43236b6cb651cd777805d46821527cdd8c607d4eb", + "size": 44943 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-08-20261030.log": { + "sha256": "f2bb3f2eb2f83d054e8d2191f33d110d010d32229893c9257f4d9bda60b94970", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-09-20261030.json": { + "sha256": "cee92389b124bfbca5c1d05a1c3cd1467b2986ef9a6bef7248537dcc3ba81f93", + "size": 37067 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-09-20261030.log": { + "sha256": "44be06be3599fce7177d1231c99727f49fc5c8a55ed0a2731a68f8754c793cde", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-10-20261030.json": { + "sha256": "b70a073249ebb1d2a688c3a6f5f068519f67beef0423e5179f966bc93e866138", + "size": 46282 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-10-20261030.log": { + "sha256": "aa292d475c5da1f78331b8110200db94bfdb297ffe373c494a7bb9bbb77cdb7e", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-11-20261030.json": { + "sha256": "0b35669d6acce8ac74eb247f50d00c36a944b274020be486bfd49821eb7abcc3", + "size": 55740 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-11-20261030.log": { + "sha256": "c03dd1dad77603e8e2b1d7dbe1305b0dcb98b24570e441c9f7a9c6c190e56a04", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-12-20261030.json": { + "sha256": "bef506c7122cdd23faccdea2f9631b32c7dbc89fbdf961313f822c9aaee85c7a", + "size": 45788 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-12-20261030.log": { + "sha256": "c2b6a2ac16c85198e0fb4d879da56c3ded8623f75d1859b816ec22c189da31b6", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-13-20261030.json": { + "sha256": "88e9bcb621608f17e1c371707074425eab1e5e60a56c7999eced6cbdf394cb65", + "size": 34400 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-13-20261030.log": { + "sha256": "043fcff3da065db6a559d8c20347807f1ab8f3d569c4ad54d8143f5534c1924e", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-14-20261030.json": { + "sha256": "e6b14ee3f5a53b3bd0d4b6c54b92db8fc78046dfe82ed834985bccc8bdb56c64", + "size": 37468 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-14-20261030.log": { + "sha256": "baf7e357a4d90469dc7ecfddc6d03d9c9c0c67db636062388d0e9176f0dec40a", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-15-20261030.json": { + "sha256": "7d3111e619275c66f4adad6cb420171e1d1c5ae80401b38288dc830874f313de", + "size": 35516 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-15-20261030.log": { + "sha256": "657358b0a89817039e39fa2b78a50e610801fb8596195ab8e8fc9390fde9667b", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-16-20261030.json": { + "sha256": "03782151798a8c423bf3dd80c0381749060f14443fa18573990d0a52cece5bc3", + "size": 23241 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-16-20261030.log": { + "sha256": "2826634e52c103bbd25b4e02b47d8498bdfba8565406060bfeb3869fb93be615", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-17-20261030.json": { + "sha256": "a0b54e31b9726f209c2972852b2ed83fc504600eace0db69d14235ef16e91e0a", + "size": 35609 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-17-20261030.log": { + "sha256": "fc925448dd3aff2d405a9beff9fba34dc23562412ee056de712aa1abb09cb3bc", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-kv-store-grpc-20261030.json": { + "sha256": "be9932537f63a21db0eeba1dd769a381ed12d7a4062fe46409a0de6b39075852", + "size": 27142 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-kv-store-grpc-20261030.log": { + "sha256": "d9fb7166a1f8bcc60aad542b8b86712848b7cb5696dff430780e1117d583722f", + "size": 285 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-llm-inference-batching-scheduler-20261030.json": { + "sha256": "dcec6c590dcc55800de1ef96a5ac0565a941be4bfcaae4b5996392a832851295", + "size": 27295 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-llm-inference-batching-scheduler-20261030.log": { + "sha256": "b962b46cdf1559c812f71b51a02c5d920cd60ceae998baf247023225e4b8063a", + "size": 323 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-log-summary-date-ranges-20261030.json": { + "sha256": "31de4194d1734a7210a35e87cfc0336575d0dc61e7ac1b2b7e04cbaa2537e075", + "size": 27223 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-log-summary-date-ranges-20261030.log": { + "sha256": "71ef8fe39ca7a61b6e463e2a628851e5c8374db5f6fa49ce09ea0ca5bd78937c", + "size": 305 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-merge-diff-arc-agi-task-20261030.json": { + "sha256": "702a32470ded74f31aac195113a429e265de167beec4bc0d0105c516cedc5bea", + "size": 28941 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-merge-diff-arc-agi-task-20261030.log": { + "sha256": "485cf057aa2901b80be705f4fc14fde437116c555a7e469b48e41ae300ae462c", + "size": 305 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-model-extraction-relu-logits-20261030.json": { + "sha256": "bd98b06439274f90039f45dd6707fb61cb4f58f02e12b4e9066054d480912d3a", + "size": 27256 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-model-extraction-relu-logits-20261030.log": { + "sha256": "7fcfbc9a4d1e0abbca49b7f69a23869451e7b422aea1df7ef82435a669b79a98", + "size": 315 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-mteb-leaderboard-20261030.json": { + "sha256": "19eac1bdc69f1a74dbe0d0897e2bb1aeb7c312b50ee21299bdc9aa721290140e", + "size": 4956 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-mteb-leaderboard-20261030.log": { + "sha256": "59cd2d4f75ae03ce35c8eb3c888497d7d8e61ed52cfccba8bfc12e726cd1546b", + "size": 291 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-multi-source-data-merger-20261030.json": { + "sha256": "38112c9c1650892699c571ef0d97074725b0f4fa0f1e1711340acc05f8ee8857", + "size": 24886 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-multi-source-data-merger-20261030.log": { + "sha256": "207d470f626becd33ab93ed60f14717571ce4553581f4da8e64914266fe6a67a", + "size": 307 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-nginx-request-logging-20261030.json": { + "sha256": "b7334855e18b0beb942d1c6a8fbae1b60a540377a57ec00e720a0f6ea6b032c0", + "size": 20021 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-nginx-request-logging-20261030.log": { + "sha256": "1c14a54b411d4b1e9b8dbcae91f0ea3082ba3e56a9ca4364db0215182aff15cc", + "size": 301 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-openssl-selfsigned-cert-20261030.json": { + "sha256": "28c5d1ddd3499b7ec8a1df2a7db46bbd60d17d7c2dc6639c06be3c1d982f6dee", + "size": 27222 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-openssl-selfsigned-cert-20261030.log": { + "sha256": "930b31d6ef2d888cc93e288e6d5130f245521d486f41584c449531be2281d5a5", + "size": 305 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-path-tracing-20261030.json": { + "sha256": "1e99fa743f588514d4d15d397341247f118eae2c5e2f30faf8bd1781d3a64e31", + "size": 12442 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-path-tracing-20261030.log": { + "sha256": "ecc99b5200e29e418a2e3347890268de35ebbdd9d64f7ba2e3f8d4709c503759", + "size": 283 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pypi-server-20261030.json": { + "sha256": "cfe6f68e746db6cfc5bf7873a49af2d0e3e89ff9591114461d1f225c5175be80", + "size": 19940 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pypi-server-20261030.log": { + "sha256": "54a145583a68b9e599bb1e1473fad02ee490ce2776e5a9811ed915a7a8561af0", + "size": 281 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-cli-20261030.json": { + "sha256": "d5e1306947c07e0db6d4d1f5ce7236a9be7cc72f089d4796e233ed5a5e43d721", + "size": 26510 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-cli-20261030.log": { + "sha256": "169b510b9e2e1ec5bb9eb78a11b9484ff26c26f0ef1bde1b81f76ee92e874e0a", + "size": 293 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-recovery-20261030.json": { + "sha256": "874a0579bfd137c5c5ee6cf6741598cc629465e16567d957745b0a622ca7dc07", + "size": 27099 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-recovery-20261030.log": { + "sha256": "c8e672ec927ff4583e4238a79245e6a1039bf266e21486b94cbbb40e6d28a396", + "size": 303 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-qemu-alpine-ssh-20261030.json": { + "sha256": "9287c434ae61bea60ded0aee594ece8a251a5bb8be69db89497b4f21b926d9ca", + "size": 11880 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-qemu-alpine-ssh-20261030.log": { + "sha256": "c7f956ae5694111b73e8775d16ccee644ef29acba1e0cac880d9febeb1d9290b", + "size": 289 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-query-optimize-20261030.json": { + "sha256": "b2e9089de82d0c6f52b5f345c6de388955b8852a37d809162a9a175aa8c25015", + "size": 19468 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-query-optimize-20261030.log": { + "sha256": "867eaf7a9741730285a6a402798b600dbbf4953a39b825e9fe08e54740782495", + "size": 287 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-chess-20261030.json": { + "sha256": "64840e42fb8b79e179382b2d8d31d89b836a6856bb93c6d618ce12697fd3e487", + "size": 27127 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-chess-20261030.log": { + "sha256": "b3e3d6ee67db301d1643caf17619ac4d3600bf3a6d721744b1da817fab112e62", + "size": 281 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-log-20261030.json": { + "sha256": "efb5503decd193c0b0edc45142e68b6182fadf8662f3a197ee23ad928681b82a", + "size": 28829 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-log-20261030.log": { + "sha256": "47a5c49587af97c2e4d38549c44951230fa243c9d9fc8cafd22da91599573fc4", + "size": 277 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-schemelike-metacircular-eval-20261030.json": { + "sha256": "91a8c4c0b5611c775b5c122414695efcf5c722d96dbf5afc21b2ed91dd050734", + "size": 27263 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-schemelike-metacircular-eval-20261030.log": { + "sha256": "c301e88a06f309cd1fecff65d012ff13ddfb7306835c8494c92cb881ab4836a9", + "size": 315 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-pipeline-parallelism-20261030.json": { + "sha256": "60fb09e7a9e852149551c7ec2930f4288785c6b4dc5451d486dcfc509b0c0d37", + "size": 28958 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-pipeline-parallelism-20261030.log": { + "sha256": "2080747e90ad37305c919f10750c5f1fca66da973b26b7d0d7b3281d5ec78591", + "size": 311 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-tensor-parallelism-20261030.json": { + "sha256": "ce23dc2c2eae0656f0a93120f2cf3185d383753534d7dece3657dec23bf68ff0", + "size": 29093 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-tensor-parallelism-20261030.log": { + "sha256": "ecd6d4adeebebd755ebc76bde167febebd5126ef6a343b99523cb7596ef98202", + "size": 307 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-write-compressor-20261030.json": { + "sha256": "a0856e05c54fc115e26f9272f30fabfd821cc9848c09b49dff5f7d319da542e4", + "size": 24368 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-write-compressor-20261030.log": { + "sha256": "e17533d12d8aa0a952500e86c672d3cd1225066779bb89bd1cb2982bfc4fa5a8", + "size": 291 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/image-cache.jsonl": { + "sha256": "523c0f267f8523ebb27d8e8fd0f0743811d6ebf4ef2da113cd654b75aae203fa", + "size": 36438 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/image-downloads.log": { + "sha256": "80a748d4f4a7e4dcb7dc5fa4c68688c76077d31b0fa1e41a1d72b93bf10ef0d0", + "size": 86844 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/initial-docker-images.json": { + "sha256": "449d13f34e0b28d37629f60676696304188487bd8f7361f38ffae2ebbe0b4f60", + "size": 465 + }, + "jobs/deepseek-harness-dev34-20261030-orchestration/registration.json": { + "sha256": "4974b32e67be7f6c7f61829c6ae9887740822cc876ad97cf5334c75f434a22d3", + "size": 91438 + }, + "jobs/deepseek-harness-dev34-model.log": { + "sha256": "d5fa93b30e037374db90244a62ea568b98a828fc46f5b1d1a0b337646f5d81cd", + "size": 3043 + }, + "jobs/deepseek-harness-dev34-preflight.log": { + "sha256": "c7a56da3a74aedc81ce20753c50c301b9151b3a1583e908902bb216f24529247", + "size": 1428 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/.anonymous-user-id": { + "sha256": "0a534b39835d68b605752ba83403f706bd17314266af0136c2fb56d62a183e03", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "370bf40fce790264a080d7b6675dbdf8336e21321299312ab1f003dcacff91c8", + "size": 706656 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh-events.jsonl": { + "sha256": "fa697cfae5ae7fee611834c6c2147bebdb7c7338c6f8d24f19c93442d646a168", + "size": 793575 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/process-supervision.json": { + "sha256": "9fd361bfccc68573302510b243fe0f8da598f5fc1449cc84b92d96fb56b87c41", + "size": 76 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/config.json": { + "sha256": "ebc23ae87749b2c7e0f9d9e53699da2ee3bdb0bd2a85e9d2520a085ac8f8ce36", + "size": 1550 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/exception.txt": { + "sha256": "8790764e6634a5aa56b79a705a9cecff562a1f592e7166fbcaee744ca5a54983", + "size": 5059 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/lock.json": { + "sha256": "2f56d36f8c0a6d7ef7292fbfceecd52d8275af182ffdecab48d0863bc0a3adc5", + "size": 2107 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/result.json": { + "sha256": "91fb7759d91c13281d4753262b4f9d41e902b877d9ccb286980e3ea5263d0b76", + "size": 43972 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/trial.log": { + "sha256": "851272872793251def5408ad063a0e1c13238d127d869e6fd26fa82651195631", + "size": 12710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/attempts/1/test-stdout.txt": { + "sha256": "f24113ce580d20e8dd656f19b4cb92bda919edbab69d2cad62788812ed8c3670", + "size": 738 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/attempts/2/test-stdout.txt": { + "sha256": "d43ae9848d99042b9c361d76f8e61dc8a56cfb0178bca16ce76ad6c5b9a04dd7", + "size": 639 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/attempts/3/ctrf.json": { + "sha256": "83d4d563fc64724981f2a34893b53063fc98a40fe99ce62a26ef54094f3605fc", + "size": 43724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/attempts/3/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/attempts/3/test-stdout.txt": { + "sha256": "923bae414e71ab4aa5ac4666eb5342bd2ccb94dc547f1ad4b96c250ec31a564b", + "size": 40634 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/ctrf.json": { + "sha256": "83d4d563fc64724981f2a34893b53063fc98a40fe99ce62a26ef54094f3605fc", + "size": 43724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/retry-summary.json": { + "sha256": "2713972eb5f28c99c93cc7137ffc023fe6a9be9b49c680386d0960c21484cd83", + "size": 474 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/test-stdout.txt": { + "sha256": "923bae414e71ab4aa5ac4666eb5342bd2ccb94dc547f1ad4b96c250ec31a564b", + "size": 40634 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/config.json": { + "sha256": "bb2df1e73c2ffd8b2c52b2ab0dfd0923db5c5e86cf2725bc3c718acba6593e4f", + "size": 1572 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/job.log": { + "sha256": "3d3244e32dded0577e3547e2fb3ffdeaa813a142d9137fb0241a0ca448a9c887", + "size": 12897 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/lock.json": { + "sha256": "0e20a32ef4d76cccc1d210f228ff5b5f24fa28d3cfe45e0c06eefa3813ceadcc", + "size": 3261 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/result.json": { + "sha256": "0e63a9fad24190d0f94c18544877433ae7f350785fbbf0969ebda1862849d772", + "size": 1292 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/config.json": { + "sha256": "a706537d7c72f1b91b34c58ed0e35496240e66329069eab0ee8ae64e7717653f", + "size": 3116 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/harbor.log": { + "sha256": "b06b3bc47b5ac3d01d29364ce51f6a3aab812ef6dbef28e0ad2f9b4b9ec886f0", + "size": 1625 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/manifest.json": { + "sha256": "e5cb361bfbde60f4ebcd3cae096558dd7d60115c43089163450eb9d2fff24260", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "7d171055eb351f34dbac86751ca2150de238c06246f4413b3964b39fe85297e9", + "size": 595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "3ef371bf851127f2cffdcea198eb17fed6e90126206697b0a3dcca19a483b4bf", + "size": 18714 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "5b90890e31ad1340294e967bdbf038a6bb0fa8233b7535d6cc8fd79cbb84baad", + "size": 12022 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "daf423ba764653aff526791bbffa6774062193d1b33dea3e6981b13f140d7da5", + "size": 15056 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "e8e2a14ed71439a888bb456af42c16452b88faca51433a117429b64efd71536b", + "size": 18405 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "eaa73bce4b7554da79787e088e7c5a4715c560e32add597dec46c8960987c85b", + "size": 14704 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/summary.json": { + "sha256": "858a410e0bd63776573eafc60426e5bd2e694352c517153d7d72ad9d5f67f023", + "size": 27000 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/bootstrap.json": { + "sha256": "0c82eb159d150060290eebcfbe93607927cd4537bc16959451b2dcc648d61767", + "size": 6202 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/.anonymous-user-id": { + "sha256": "1b70d901f1108257f2c99a1c5b3ede5e189438301accfcb47f79d37f03cb6dec", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "3f3d77d40b475c4020f389c06a32db7c79d8c4d8727f54e1cbfdfad2fab07ad9", + "size": 478273 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh-events.jsonl": { + "sha256": "6f96b40fb8f18b878b20bb161e816601b4f7c2197b0409f3a583ea0fd1d3f7bc", + "size": 536870 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh-result.json": { + "sha256": "37aaa6c503dc43bc377a41b736aa728ec8b852415d03b6e0328c80586a835b82", + "size": 1194 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/config.json": { + "sha256": "dc82b284a28e26cf9515ff9b1cbb40ce99a679a37618828a3992d894882add5b", + "size": 1520 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/lock.json": { + "sha256": "008b7fc73cbcffa08d5de751daeaa4e1e1349e655e1a6c28a409f4d76b84179d", + "size": 2097 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/result.json": { + "sha256": "a11acf668fc25f24e0504f5435a9872401145fc601ac94645f9dab6e41704725", + "size": 14415 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/trial.log": { + "sha256": "1d1e3c54df9c39b7658e7eb76fe1730d108386cd6c9a0fd9fda4862c4cc77cf6", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/attempts/1/ctrf.json": { + "sha256": "fbd391f554b1652f97518fca51b61c396f0a7db7fc59da8639cb6976a7354aab", + "size": 4260 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/attempts/1/test-stdout.txt": { + "sha256": "9de6887f99b6567a850f9455ee552bd5f63021d34a9f03e4979c0aff98cb4ed3", + "size": 3526 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/ctrf.json": { + "sha256": "fbd391f554b1652f97518fca51b61c396f0a7db7fc59da8639cb6976a7354aab", + "size": 4260 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/retry-summary.json": { + "sha256": "1320ddee25ebcdc4bd943a4141e9e19b61116e3ae973b9166f889a73b559c1a0", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/test-stdout.txt": { + "sha256": "9de6887f99b6567a850f9455ee552bd5f63021d34a9f03e4979c0aff98cb4ed3", + "size": 3526 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/config.json": { + "sha256": "2c821fea856ad705b6c6057aeface0701a11dffd212a9eedb034c220d76e48da", + "size": 1552 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/job.log": { + "sha256": "1d1e3c54df9c39b7658e7eb76fe1730d108386cd6c9a0fd9fda4862c4cc77cf6", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/lock.json": { + "sha256": "618c5205c3825d302d1a41984033e4dfd44f6b06e3436a992cc5b53f7ec0330a", + "size": 3251 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/result.json": { + "sha256": "c17d06c01c064d9297214bc54a78e8716348a04b66d72adc412262d932214a89", + "size": 1140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/config.json": { + "sha256": "aa3f055b4e578b91869d2d6cac27ff50ea8f1a5f597ff0b3a770f64b21ac6a2e", + "size": 3096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/harbor.log": { + "sha256": "f0e649488d83d1885135882f47f98519732dcdd1ffc6f30ccad849646aac1597", + "size": 1268 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/manifest.json": { + "sha256": "e5cb361bfbde60f4ebcd3cae096558dd7d60115c43089163450eb9d2fff24260", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "63599de68cfe2a0d57f7784153a726afb5fe89dfe92775b79ae81c8bd72e4af8", + "size": 585 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "6e155faa6dbdb6f91941ef64e46fe762b45331fa6b2b8fc1fc407555a000cb04", + "size": 18704 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "7f8e75ebc7cef72e09371e99bbb0df1a009ed1790edc0ac02ebe4b686a44fc67", + "size": 12012 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "56b50ceb0251622d6b3f83e9ff8f61558dbc902efc303a567138eebc18bbf2be", + "size": 15046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "18b7cd90260b38e5ec861ff7a3caf536b157d91a12b02d349c2bb464c4dca8b9", + "size": 18395 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "d95510ce339443956e9b6d6a76cf30b914b98245a5a2169c77622fc41ce06db8", + "size": 14694 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/summary.json": { + "sha256": "e6c42cfc744691a2812eaa71d2fcb28c9a4e4f55c90eb891e6966e82a1dcc8c4", + "size": 7672 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/bootstrap.json": { + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d", + "size": 7379 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/.anonymous-user-id": { + "sha256": "89c5db2ffdf99af5bf30b6a2db1367b999a02e4606a547f82023b0ceff0bdcb1", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "65325f1b69839499d1b3cc7a2b58b7928ef8333b5b174981c095c2a633ead898", + "size": 324883 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh-events.jsonl": { + "sha256": "aa213998b0adb6b57a19d2b73a70fd00f9070c201b82dc99a4cb600276efdfe8", + "size": 362964 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh-result.json": { + "sha256": "7d496e4257674c36e6efdf3f9091df51eea04400f2bfd5b28814a0e9c17dd88f", + "size": 457 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/config.json": { + "sha256": "c52fa929bb256cd8055667458304e28a83bfef8ec140d679f3a99c8e39edf474", + "size": 1511 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/lock.json": { + "sha256": "7f1cc815cd5bdc8c250b4f99fa83dd8b3e53716b462340bdf17433e9ae92807f", + "size": 2094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/result.json": { + "sha256": "1a55e19d4d065a03cf0e8463a23fcfb2f01290117d32a75d2d740dfb51853d5d", + "size": 16009 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/trial.log": { + "sha256": "e27f0b1670d54753b420f7122bcba8cc75e11d1f797cebfec787ca5d6ccb57be", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/attempts/1/ctrf.json": { + "sha256": "f1c2877cb582943996f32d31c56a54ada9fb73b10ab45fb4ba9dc469a926d98d", + "size": 1467 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/attempts/1/test-stdout.txt": { + "sha256": "e0db42279d971b35469d51ac5e20f9c3de9ff9143219300819e47b220ec829ad", + "size": 1914 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/ctrf.json": { + "sha256": "f1c2877cb582943996f32d31c56a54ada9fb73b10ab45fb4ba9dc469a926d98d", + "size": 1467 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/retry-summary.json": { + "sha256": "c22fef9936b4b5e2f7c3160ab737580603c2c5758819750ea79f7453d11d7bab", + "size": 272 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/test-stdout.txt": { + "sha256": "e0db42279d971b35469d51ac5e20f9c3de9ff9143219300819e47b220ec829ad", + "size": 1914 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/config.json": { + "sha256": "3d586945d4223a97b88d1dca2397b55808915973e21dc15361d0b8c3bd3f42bb", + "size": 1546 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/job.log": { + "sha256": "e27f0b1670d54753b420f7122bcba8cc75e11d1f797cebfec787ca5d6ccb57be", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/lock.json": { + "sha256": "662cdc20bcdfb888876a518d460e4ff9ad6757424599cf654e13ad712b3ca750", + "size": 3248 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/result.json": { + "sha256": "be21a4e2800d1312002fc491d4f5ae9f9d1afd2cbba10f049310930567cad07f", + "size": 1134 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/config.json": { + "sha256": "1b0f9b61f0e331332ede265d919619df18bd37dcf6f53280ab5c8ed0d570129f", + "size": 3090 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/harbor.log": { + "sha256": "9abe356f57da5becc2157db141220256a6b90b98f799b1ec1f68698e066eaa3c", + "size": 1262 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/manifest.json": { + "sha256": "d67acc20f268c504f07405fe5a4fb004985fe5c881127e6ee94147feebfe3006", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "b62d65f2fe0857b8eaf3618394590a11584372b395ce0cf42dd72c6fb0fa9416", + "size": 582 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "5b089ffe394258514fe66e6af3072f07d2d7f1a23621b43f2fdbe1d64cf18c6d", + "size": 18701 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "0952c3f339b16ee2bf1660d1ba811adc67f70bc52ff74bd58b29f00d01a86a3f", + "size": 12009 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "efce853b407e99a91b7e6b87779832cde56bd0c676a2d644e5a0bf43f94b1695", + "size": 15043 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "3ea7deae8d08e0e80912271d223621aff1646a97049ce30ced52d806b65633bb", + "size": 18392 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "89608f6d278cdbcc6cea9b262f75e135bf015055f87fb1b6408c7a42001c5dfc", + "size": 14691 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/summary.json": { + "sha256": "828b949d9ca076c7d5e98fd101261aff6da70e7515e209a40cba989b9938a02f", + "size": 8971 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/bootstrap.json": { + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d", + "size": 7379 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/.anonymous-user-id": { + "sha256": "d93c5840de0ef0b882fa89c7ffcea710aaac517d63fc6beb246c9327771da601", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "3342651204e9793e86a9d93e7522b0090072e1eceeb28e3890326602207838ba", + "size": 546948 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh-events.jsonl": { + "sha256": "9f3d9966e0fbee53a604346d14d3e6e147795087c3cf652a95007531a51a18c7", + "size": 615945 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh-result.json": { + "sha256": "d680af55446a5b5dcdb7d170a61ab50cdf9a2bff4fa5d87da4d60c0fb14dd604", + "size": 696 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/config.json": { + "sha256": "85cd257bfba429e473220c9a7bf0ad7a58de5d87807bb35015989def9d2c7677", + "size": 1514 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/lock.json": { + "sha256": "435e6da3985c17ff096216063f0037fe8ae856d191b09ee4cd3dd8c5c333c166", + "size": 2095 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/result.json": { + "sha256": "09d2038486503d272f70461f6d21f39f905ebb5b73bb2a35b2c36d7cbb8de5c7", + "size": 16020 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/trial.log": { + "sha256": "1fccaf25940b4e134c0dab92837454fcc8327afa6eb5b03b7fe60cd915343a47", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/attempts/1/ctrf.json": { + "sha256": "864094d742d1e1c4fd90afd1a0a8d787a95e0a20472b4c5ead2d3223c25a6f3f", + "size": 2542 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/attempts/1/test-stdout.txt": { + "sha256": "e9157af437d576f9d81e98ee3314eb6e702832a842a645a62340305541a637fa", + "size": 1816 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/ctrf.json": { + "sha256": "864094d742d1e1c4fd90afd1a0a8d787a95e0a20472b4c5ead2d3223c25a6f3f", + "size": 2542 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/retry-summary.json": { + "sha256": "ef3f73839e37b98d0d43a86d1e73e0075189596159ad1f8c534b87a60dc40a80", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/test-stdout.txt": { + "sha256": "e9157af437d576f9d81e98ee3314eb6e702832a842a645a62340305541a637fa", + "size": 1816 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/config.json": { + "sha256": "8ce9952d8495ea643fcf069341e2f3c35ce11340be087010cb6cfbc4bc75fdbc", + "size": 1548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/job.log": { + "sha256": "1fccaf25940b4e134c0dab92837454fcc8327afa6eb5b03b7fe60cd915343a47", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/lock.json": { + "sha256": "f0a0d9fc4b624c222d8da4a82e0d6b017af1c54ad89757fdff024f7fae4428b6", + "size": 3249 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/result.json": { + "sha256": "f69208a85812ad437a2205de3a8724eec88501d3096e89eb085f789d316df4c3", + "size": 1138 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/config.json": { + "sha256": "9f5463c2fe65e195928ed9eab0033d2f8fd0f6f807f98667f321f49a1c7eba10", + "size": 3092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/harbor.log": { + "sha256": "161049b3f2667749c3050c833c24f04ba073c89ffa03315343c16b3db12ff88a", + "size": 1264 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/manifest.json": { + "sha256": "e15b80c308deaea2c9d073f59599096b6379730a25029b0e6dc4b306300089a6", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "a3bb8b965913d747526daa19eab3888a2485583d8e0c08394fca0a1afbb1b0ed", + "size": 583 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "2afe7649d1c04580c3ec8ac538ed1c6a5fcd3005b5078ae43a538d96ff3c6fe1", + "size": 18702 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "b860e0f7e0cef9a6c293a98d3daba6ecaa17fbcc59cdf9ef834477a615ea646c", + "size": 12010 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "e82fa813b552608f7417a8505a8f99798521a5be873738c0a418f2f6dd0e1044", + "size": 15044 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "166af1928d65c0d185cecd41ff98484854845d4becc101cff33b4c6fbac1060b", + "size": 18393 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "cfca50d8068021fc3da94ec63e161f4b795b1fd4a868c1ea5c136e062daec369", + "size": 14692 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/summary.json": { + "sha256": "f73d2a0fd55671bf8ae6c4ac589e538a1a9678d05a3b83eca82d6976e4c29c46", + "size": 8971 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/.anonymous-user-id": { + "sha256": "0cc095a22a1e7d7829131d8ea4fd6669fbe20a7ab40c245bda225a7e03c57434", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "21c0fc5019821164b9ec65d1cf44583e2d4809767e8e02966e9e8a7024208f79", + "size": 72648 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh-events.jsonl": { + "sha256": "9e7945012332239fc12fd55667a630ff199e857a73a0a63ee71df31a282e1ae3", + "size": 81877 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh-result.json": { + "sha256": "36f7857ff2ebf31c1927f54cf472c7b267f8449ce2ea73b478dea1585cb0e115", + "size": 437 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/config.json": { + "sha256": "0bbf00ee799592ba88b8c128e358ae296c951c53c9f16d3fb0fe77fa22892b80", + "size": 1526 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/lock.json": { + "sha256": "93d3afb553a013744c6bd21b7ca58bb7a251fd25361881a6293aec9ee01a0f08", + "size": 2099 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/result.json": { + "sha256": "ab0c8d2685ec1064d386684942774d41c013ad649dbc78a1b9f4a6e2e3b2d3e9", + "size": 36457 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/trial.log": { + "sha256": "35e23ca70bcd07cedc9b99ad17bff055b511f17b357546278318aa1a1ec06593", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/attempts/1/ctrf.json": { + "sha256": "89d39fe271aeb0ac2596c2903eb7f0c3a43adc9d4f84b4e4fafab636c8df28c2", + "size": 4082 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/attempts/1/test-stdout.txt": { + "sha256": "f0a8804f6456d9a566e33de1bb448baeebce37177b8474f9b99401ebf4d64fda", + "size": 3044 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/ctrf.json": { + "sha256": "89d39fe271aeb0ac2596c2903eb7f0c3a43adc9d4f84b4e4fafab636c8df28c2", + "size": 4082 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/retry-summary.json": { + "sha256": "ba3ea17b9be77ffc531fb345c98ed7cbd38c57aaed9c202bc6c3b6014cd37c95", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/test-stdout.txt": { + "sha256": "f0a8804f6456d9a566e33de1bb448baeebce37177b8474f9b99401ebf4d64fda", + "size": 3044 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/config.json": { + "sha256": "74fbf902793a22ebb6d4ae97cbb9255b1e4b1bc22138f1b67ce60b0bc7e4db97", + "size": 1556 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/job.log": { + "sha256": "35e23ca70bcd07cedc9b99ad17bff055b511f17b357546278318aa1a1ec06593", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/lock.json": { + "sha256": "218caa94c13bc6c1a65044ab46ed0b03fc2b304e5f931a394110cc76a069f5be", + "size": 3253 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/result.json": { + "sha256": "c16d0469f743125a34676d6582a69bf37d5cad472df7880008d8454fec8dbf43", + "size": 1135 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/config.json": { + "sha256": "9d21a9336976fe82458b423ad8bd14979a087d1badacea94e5570bf97e94152b", + "size": 3100 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/harbor.log": { + "sha256": "574a3dd62ea6b7dd4bb6249bf3ace4e8966a5382c77cf95b79ed02ae1a66c1a4", + "size": 1271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/manifest.json": { + "sha256": "deff376d7643827846095b81ae44f57092fdca37913c59372d09fce0f72b35c7", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "a1a6a5ebb4dab44608cf4b1f9c28e5524b298c96b6c2e1faa6076dd61728d3cc", + "size": 587 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "a37ef56d163ad813a467f61d2e00d823fed85f0c01a7baa82fed463c49b49da3", + "size": 18706 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "c73af2331591b7580c5c0a90675c89ae5020b10825d9a4276be4d3ed4c5175f4", + "size": 12014 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "1e012e08200021fa9ed218a66e3bb08ccf475e359b2fe720facb69640e483a7b", + "size": 15048 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "e44dfe77de954f05621c84ad8adde4587e1d8d62ceed65ebb8db221872a992b5", + "size": 18397 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "518ac22b370cbbfe18c590fc74df7cf22524c7c84768512fa5a5f124c61daf79", + "size": 14696 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/summary.json": { + "sha256": "f5ce4960580f28a0af4c29aa3ceea8bdd3fb33406fa9d3405d2a33762f1da5a4", + "size": 25046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/bootstrap.json": { + "sha256": "4e51e6b6cda713751b4e5476ece237e7a377c6bba5e5893b04a4cac95a2baaf1", + "size": 15851 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/.anonymous-user-id": { + "sha256": "89aba47e3e136f3834428cbfa01352e429285b3d617f6579c9c11adcf853c6c2", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "fd2e87735db62d1e22ab07449388e0f22bf4616d6a009d750197bd8cae1a7839", + "size": 987635 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh-events.jsonl": { + "sha256": "60982612ae2d7b8ad86c2ddbe04afb5486971bcb627b9d633860edc1eee91310", + "size": 1111745 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh-result.json": { + "sha256": "6836e4f534c330c8d54e9b2fe4dd8fc5c3ffbde360fe34d8db47eb203bc34c42", + "size": 389 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/config.json": { + "sha256": "b91f0f0eb8257465548f636468b8d16ee1193a565105f3edbee7f8fecf6a03af", + "size": 1517 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/lock.json": { + "sha256": "999e7b4b6f21fdce39125a9f1d5a869b4fbd2737fe9adcdf339ff0513753bd36", + "size": 2096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/result.json": { + "sha256": "0b575ea99418eed3108dd653f769b67d8bd58e8e7e9b790f76551114c72d01cc", + "size": 28050 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/trial.log": { + "sha256": "1fc9d5015be7b6bb16efd4813c609102f60bfa41ca90eb32ddd66a15b6a90760", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/attempts/1/ctrf.json": { + "sha256": "58fe384eaa09ad0a1b7ea5a86c56a61be392c02b97b1d723132b606e1dcb8a0d", + "size": 1435 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/attempts/1/test-stdout.txt": { + "sha256": "56a10f48113ce0398a1a03d05a3c2c68ecc8bc257732c9370a7d5a316b602a3e", + "size": 4487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/ctrf.json": { + "sha256": "58fe384eaa09ad0a1b7ea5a86c56a61be392c02b97b1d723132b606e1dcb8a0d", + "size": 1435 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/retry-summary.json": { + "sha256": "9165e04d7545e0e6357a7851c7d6454771791c957049b5f5fb7ef16257c7a13b", + "size": 271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/test-stdout.txt": { + "sha256": "56a10f48113ce0398a1a03d05a3c2c68ecc8bc257732c9370a7d5a316b602a3e", + "size": 4487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/config.json": { + "sha256": "1a79f6172cbe836ebf67ad332b1fc17e7ba4d9428d9d7c02c53fef52d7a03e4c", + "size": 1550 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/job.log": { + "sha256": "1fc9d5015be7b6bb16efd4813c609102f60bfa41ca90eb32ddd66a15b6a90760", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/lock.json": { + "sha256": "4d04ddcfc2470006bc7f8166d9bb1efffd3a371f3821b0e5f1c46d0851172d6f", + "size": 3250 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/result.json": { + "sha256": "21411ff007a047eeb81893625dc266a82560303c7388cb5788643cf6b498a343", + "size": 1137 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/config.json": { + "sha256": "05a46047d538cc18f7433d60ee9341b4dc871923402dbddeb2b514b3293398ad", + "size": 3094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/harbor.log": { + "sha256": "b935cb252cf936361548ade619ca6cf6cb659ca8a4297a6d56bf22b8905f6b9c", + "size": 1265 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/manifest.json": { + "sha256": "144d39b5e20bf5333fa90dbcb759eb8a8b78833460b856bde474236828cb8bc3", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "59db6c2586d048eeb570cf22db5bf7d7e39b8d40aabb44f13330e29fee3ce0cc", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "fdb1a527be76f5761d4f895c1e9fac7d779ed12f38bf67dec9ee805c649fcde9", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "8d4df943d83791005f263d83f16b784b425c2f6550c81a24f9105d8310c210ac", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "804070511bf346f99e92d3b53cc97666d5aaea7789e25ebad45c3d9e365d9553", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "c8ebe1cb32e795329395876c2200cd62ee20a54c73ff4cb66dc4ada8cb1c317b", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "304c66a66302d241e3a0170d690b94b0cba7c6cdf725e045c6995329a311a555", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/summary.json": { + "sha256": "d759da4cc0c20798bbd1e5c900f2520afb411f38c191665051b0e0fc6ac351aa", + "size": 18454 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/bootstrap.json": { + "sha256": "4e51e6b6cda713751b4e5476ece237e7a377c6bba5e5893b04a4cac95a2baaf1", + "size": 15851 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/.anonymous-user-id": { + "sha256": "b70b3bfbb0039f8f70c59d98b88b4d6dd7b00ce5f24c436dd8becbb1a7049307", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "5a70d4084c0d98890da1bcc04e73e02192f2d7722f700be42042f109b9539b7a", + "size": 443664 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh-events.jsonl": { + "sha256": "53339f995859d05413b8aee55ce34f7b46871518ceca70d1385a39f4f6269c46", + "size": 498128 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh-result.json": { + "sha256": "e805b5920a0c19181dec73f55cb40d1c70d0fdfb31a016f7fcc4b985cb57e190", + "size": 451 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/config.json": { + "sha256": "6c38de04077a9c5848b00e38ce4364a730507ac7e83c2312a386c15bc3ce9996", + "size": 1529 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/lock.json": { + "sha256": "1f708088501d71c2d8b5171a27376606b031b6fa19c54aa97010881953e73456", + "size": 2100 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/result.json": { + "sha256": "548dd8306a610d37e299f879e648ed69405af406346231be18f0c34a680092c8", + "size": 28082 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/trial.log": { + "sha256": "2219d40540a66bf673e0b18e55862165bfae37ff7606fc730db3af15a0cf52e2", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/attempts/1/ctrf.json": { + "sha256": "454f52b47c83a1a232c3e461df6d226c19d109cd4c24c6f71cc020fc360b0f26", + "size": 1448 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/attempts/1/test-stdout.txt": { + "sha256": "ef4690316660e8cef4af73912d0b4140e3f96212795cc4770bcfbab33eea1289", + "size": 1324 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/ctrf.json": { + "sha256": "454f52b47c83a1a232c3e461df6d226c19d109cd4c24c6f71cc020fc360b0f26", + "size": 1448 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/retry-summary.json": { + "sha256": "b2ea13ea8639e2f51a9d0d4a7e1150301d022cb83f09589db8a04ed8581a5e0a", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/test-stdout.txt": { + "sha256": "ef4690316660e8cef4af73912d0b4140e3f96212795cc4770bcfbab33eea1289", + "size": 1324 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/config.json": { + "sha256": "2c76df44b2b77efd0d69e29b972f536acc3bf54d0e47c9174285da95826c6602", + "size": 1558 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/job.log": { + "sha256": "2219d40540a66bf673e0b18e55862165bfae37ff7606fc730db3af15a0cf52e2", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/lock.json": { + "sha256": "61e36c9ca1d5381199ee6a02b8a7a58e4d4e436af96c9b700f688a7498ab4f3f", + "size": 3254 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/result.json": { + "sha256": "04021c08780df494c190eb9823e746654b43eec7a0f77ff36583e4b048813dac", + "size": 1141 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/config.json": { + "sha256": "c8c45654330a62fbbee86de53236aa014930b6722b58677d8c5beed4b4d561ff", + "size": 3102 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/harbor.log": { + "sha256": "027b94f9859fcde1cbe3693bf14d2e94070f2670d63691e2d3dd67aaa08ebb6f", + "size": 1273 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/manifest.json": { + "sha256": "deff376d7643827846095b81ae44f57092fdca37913c59372d09fce0f72b35c7", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "a928b676d44483ddf5d5f994af800b49fdae4419d554b5a6e4b07e879308c36f", + "size": 588 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "db9bf472f263bdcb42154f2ecc61bd911d15ae49fc9e1a8df9b012c12e598553", + "size": 18707 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "d8b92cd6f4b7d42ea2cd956f1be2861c15dcb8605759a1508f7b7c717db8b682", + "size": 12015 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "0b8fb7dcd3cceaf3b4cb52b2ca7d7555c29f13133de8f3040cff59c6f28ded53", + "size": 15049 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "7ea390ed04a8bd34aa99fecf23c80fdd94fb0d64e71c3979b51e79972ed65080", + "size": 18398 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "38633bc95b5a22b812266a093a88410d4ff828ee7f8ba8b1f280089decdb1691", + "size": 14697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/summary.json": { + "sha256": "c25ffacecc36245569ff2b80f29f1cb976ec9c5bb701b0b55c86b3302fbdfaa6", + "size": 18461 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/config.json": { + "sha256": "7da9d000aeb3ba7861d8c0fcf80a5127f9cf1245ed2a143fd3f1ce21a87d5ead", + "size": 1564 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/bootstrap.json": { + "sha256": "cf2881c5ea9a508fcf319b2a6befc60d39bd25004a7d4f3d2b194579b31fe401", + "size": 18592 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/.anonymous-user-id": { + "sha256": "eee3cea969a23f17f49987b5e7950639cb8530c0eb64894263a47b2b89bc09f9", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "b3155e5d0cccce5d47dc76c5d6b21c1359c96e3168c2757e017894933c9a3dec", + "size": 56602 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh-events.jsonl": { + "sha256": "71fdd2595bb7d4af395122260ddbc6dacfdf730d8f35184a351f098237cd879b", + "size": 63913 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh-result.json": { + "sha256": "3c6fd5406f0a007aa2b6046e959b60e013548ddb58a627d4bef43bd2bed883c4", + "size": 326 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/config.json": { + "sha256": "b65341f7cf976348eef26865edfc5d0486de40914de72ffb483f4a33c5fad0a6", + "size": 1538 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/lock.json": { + "sha256": "e6114dfdc3f300740b2cead745616ab98ac5dcfc534a03efa9035f2d29771ce7", + "size": 2103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/result.json": { + "sha256": "6790ec0f4a1f9ccb9b191af8361ae8d77bf7f59794aa2f4e04fc0238d85043c9", + "size": 31877 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/trial.log": { + "sha256": "d3ae3904b338ebc6e5323a2159c77a40de26443ed5ce7edc342ccc59cab4bb96", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/attempts/1/ctrf.json": { + "sha256": "54517b5a87fdd991a11e75746f9b43c3371add91d654d50593fe39e64201f367", + "size": 1517 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/attempts/1/test-stdout.txt": { + "sha256": "cc98ac573e8853538edc9451bbb2c12830300030b925ab658cc2c1c3b64cebf3", + "size": 1454 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/ctrf.json": { + "sha256": "54517b5a87fdd991a11e75746f9b43c3371add91d654d50593fe39e64201f367", + "size": 1517 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/retry-summary.json": { + "sha256": "f58a53fcb7b5c1ad7d8666b587d047bc7c01fcdcadf0f3a738f05e7187345623", + "size": 271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/test-stdout.txt": { + "sha256": "cc98ac573e8853538edc9451bbb2c12830300030b925ab658cc2c1c3b64cebf3", + "size": 1454 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/job.log": { + "sha256": "d3ae3904b338ebc6e5323a2159c77a40de26443ed5ce7edc342ccc59cab4bb96", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/lock.json": { + "sha256": "b4d2e6059a61e0342030ea72be9299be3de179ef1fe1f7d2f8478386905ac84e", + "size": 3257 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/result.json": { + "sha256": "bc997e9f9a66974aa6d013c02357532186f1f40af70d690d10a1dcaac87a12b6", + "size": 1139 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/config.json": { + "sha256": "79c6b95efc3d932b92f3d4f4054faa08cc1407b6d79515f16f56844724d85c5b", + "size": 3108 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/harbor.log": { + "sha256": "a8fbc8353a6d27ae3d8589eb66b0d37fd9d489a7ca2df1534e59fdc6676a6b40", + "size": 1278 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/manifest.json": { + "sha256": "49b36ec4775750b7cd22b72de420cceaead5c20a53502a75bd3f9976049bd7af", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "a2c99855c250faaa319648cfa7c16b87659fd0be5e5bd1eb6bfcd5292519b58c", + "size": 591 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "db0f22e0eb5bdf6c28b404c894170744afe0755c32ecaf18f8bb2d8cd02e7c47", + "size": 18710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "ed2ebca98bc27b5915ba6137242ec4c77819b44481b030f74139158e75acd38c", + "size": 12018 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "7951e0707d039116ab62a99b051a021991dbace05457c87bc434db2f0e86d289", + "size": 15052 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "aca6751e0661075fd64f6dbb52159c169a87ebb524605d842f29d4a395ef535a", + "size": 18401 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "011b231fe2ff36b4e43d6e98caa63d4b68133547827ebbd2513e5f1cae171428", + "size": 14700 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/summary.json": { + "sha256": "171a0ed005889d2c1890edad8b7d3e5bda9ff7dbd874e3e9a024a8290b4b8105", + "size": 21503 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/config.json": { + "sha256": "f4e71ab95323549622ada23395c1d0f2614af3dc98a0b6bbf2178cc017373c90", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/bootstrap.json": { + "sha256": "747599891fefaac25e7ab1119fee62bb88a9bf9b3e036a496c37c79d15610b74", + "size": 4819 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/.anonymous-user-id": { + "sha256": "dc18db48fa1f87ea9985795f8b4fe0a47ad92ab7868906821db9a006815613f6", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "5f500d2d9f961dd101ee8d81b4c5132bf9d271676be54bda6a2873aaf790348a", + "size": 349254 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh-events.jsonl": { + "sha256": "38807afa25b865ea21d58c186b5dc66cac08cbc4881cebb9ae0a12196829162e", + "size": 389421 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh-result.json": { + "sha256": "8791d08b5a12b7b84a3f332fb398aa3933f923130033befb321021cd38ac87e0", + "size": 970 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/config.json": { + "sha256": "ccd985846c1050baf54be8c087390c0a77ea6aaf84ba57eb360086b995455268", + "size": 1544 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/lock.json": { + "sha256": "e603be65249637893feb4c580a82ee4ba95e6f834049afe3976e6484eb7027f5", + "size": 2105 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/result.json": { + "sha256": "50c651e46529ccb383769ca6a63e5d36d37bf0ef57fd5d412eabd623cb3f0e40", + "size": 12502 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/trial.log": { + "sha256": "7787730c9ed83e53c14dd4fb3eee36324dcc6aafd737036e0420a6b996f97565", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/attempts/1/ctrf.json": { + "sha256": "0c19ca537f239adf0169fe147474c59497072367f85596d0ef0c01c34d6bb083", + "size": 2548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/attempts/1/test-stdout.txt": { + "sha256": "40dc429d0e27f09c31a68177111c6fe452d135367444fc487d8f7921d1a9e632", + "size": 1617 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/ctrf.json": { + "sha256": "0c19ca537f239adf0169fe147474c59497072367f85596d0ef0c01c34d6bb083", + "size": 2548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/retry-summary.json": { + "sha256": "809e75cacbffd614d0e49daefbc4ccf3525152b7c75a710c9d5027745400f0f8", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/test-stdout.txt": { + "sha256": "40dc429d0e27f09c31a68177111c6fe452d135367444fc487d8f7921d1a9e632", + "size": 1617 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/job.log": { + "sha256": "7787730c9ed83e53c14dd4fb3eee36324dcc6aafd737036e0420a6b996f97565", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/lock.json": { + "sha256": "9dddfcc50c6f7061e19b5cc7e585c219fefa72f4c64d1810912b9e1a08cf5ce3", + "size": 3259 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/result.json": { + "sha256": "92f5d628318ad20670e188aa18175e4f22e40df14a24694b67e8a9a158d95253", + "size": 1146 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/config.json": { + "sha256": "30999de9f48bb348a50ab8c27e23185ec6c1d6e20d39c3d830ed9d445d5a454b", + "size": 3112 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/harbor.log": { + "sha256": "1460d167e271bb54ae64524ddf064df1a2c06a692a956ecc83e8b6b38ac5a1db", + "size": 1283 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/manifest.json": { + "sha256": "e5f5ba1a234be82c7021354b6052a1717451e0de1e84e57e668823ecf88bdea9", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "8997a9c0d2dd7da5531112929ac211fe255252d358ef0c0950d38fe05a8748bc", + "size": 593 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "ab7e11c71f7dc296abf1aad432945513bfda75f8faeca177348423c0f1e3ef66", + "size": 18712 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "fdbfaeffb9b16add579c4a08c03981008be944609ea72ddb5bac2af474f404a2", + "size": 12020 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "b4d2213e0be7f0b2c142fafce3f8f47c17d440f64af25a32aff8ffec210bbd17", + "size": 15054 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "c2dfa26e6c3890c941cd39fe7539b6a554f08fd6fbaa926ed978db438456f31a", + "size": 18403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "7963cfc36aebd1884e33ebfd356104780abeda95dfbfbac4dece35a4b099cbd0", + "size": 14702 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/summary.json": { + "sha256": "dd4989fc06a0b9ef3be97f59d18e903b84fa868c53180744321b796827305a2a", + "size": 6137 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/config.json": { + "sha256": "674c18527639b66c0a7544925fc8ceec29d9a717f0595539f9050cbd29c31213", + "size": 1550 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/bootstrap.json": { + "sha256": "317eb4d1a905a9446f48890f4be51b9acf3c0e230c0770b360285bae70d579b1", + "size": 12040 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/.anonymous-user-id": { + "sha256": "3c98d8687152b89c0ea67cfaa52be035d94a6004b369abc120896f819ca1992e", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "e727ba72040e974e3a367b6ae226542ed5f8cbb6184af7efe7aacdb196a8b494", + "size": 80658 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh-events.jsonl": { + "sha256": "b74c7f1e760eb5e2ee7ad1ad323c4b0280c5b2d55e1c0007059a667a5efba7b2", + "size": 91205 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh-result.json": { + "sha256": "8344593134e8c67cc86a9b0ba1b2019f82feecacd05840fd18ec6dfea48c2cba", + "size": 960 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/config.json": { + "sha256": "81c073a31c67d1b12776902ee84d7c8339dd4c2485fd44b7cc159a0e169a26fd", + "size": 1517 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/lock.json": { + "sha256": "2fc8b7fef1156c034af25540120508eb1016250a21738357346ce2aed50b9afc", + "size": 2096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/result.json": { + "sha256": "84708abe150c2fcbb37e5dd8449d5f950dc4f6b23a23e49859c7d7f70d643d18", + "size": 22607 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/trial.log": { + "sha256": "68539bae9a967f779bf3bd7db61ab548c238af4e8be810e4153ab57119564007", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/attempts/1/ctrf.json": { + "sha256": "129c51bdcd43ac7088e15dc70652efe56a47469d89a90a47749bda8701fd1d01", + "size": 2880 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/attempts/1/test-stdout.txt": { + "sha256": "0a393eb66bc19a31f0f070f2ea377d8b0e11cb35b8130f44155a59249239d546", + "size": 1640 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/ctrf.json": { + "sha256": "129c51bdcd43ac7088e15dc70652efe56a47469d89a90a47749bda8701fd1d01", + "size": 2880 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/retry-summary.json": { + "sha256": "253bc8ac712beccdc4eae0b28c56c5ad6d00f723ee173223d5ccddee6edeac8b", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/test-stdout.txt": { + "sha256": "0a393eb66bc19a31f0f070f2ea377d8b0e11cb35b8130f44155a59249239d546", + "size": 1640 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/job.log": { + "sha256": "68539bae9a967f779bf3bd7db61ab548c238af4e8be810e4153ab57119564007", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/lock.json": { + "sha256": "f79644ddc6348a055b50099580933a84fadf711ce6007b8934551cc35c463ed3", + "size": 3250 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/result.json": { + "sha256": "07abebfde84b551324864c7cc657c3a1354fa2f5d424d0a99f7fee453c5a4072", + "size": 1134 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/config.json": { + "sha256": "14de9939fccf109fc585ee64b829d37acd26382e43b8f8559c926fc7afdc89a4", + "size": 3094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/harbor.log": { + "sha256": "276e8121c17fc75ac4b87526b14e330f8b48ad2e87ae07574d08716f777947a6", + "size": 1265 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/manifest.json": { + "sha256": "1b9a6fcc70ee8a059b5cfb1d9a1af34a34798426c27264c25924c43ceb8b1031", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "f3cb31959ea5c512704e7a9b519f330d7b6f336fc71996d686642f8756b25fb7", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "454a75ec6f6aa056c768801a7c4071fd3d8e4acae645107c068e39cd5a4a9244", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "158883d07bd40cdc968ecf936ca89cdbf7bdab0aa343f66812383223d1003712", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "3bfafe4d6b619013733c0b028c0c3a5b205005c11043806bb12d0f3e56161114", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "a510129e4924a78df37b97d8070471c32ada5a92266814fa2a3cfa22cf3b80ca", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "d857ba4f479202a6878f8ba229cdfd962b2e06d7964db2dfdfcb4827e1794df0", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/summary.json": { + "sha256": "cdb4abd250ad1f55b1bf3698024a77b3cdf9ba73f706030b670b170cf666c9e8", + "size": 14180 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/config.json": { + "sha256": "06b01c62b4aca5601ef89952b3bcec2d2c16f84c8ccf83c928a9b73a52b7fb78", + "size": 1544 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/.anonymous-user-id": { + "sha256": "6fb065b24ba784f2022494984fb5487b6672fbb4e6ae68b8d79033d9ac719e05", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "dd1de085e5675e2395d7db644f13802f065c9e651e8c1a9039aac748618a345a", + "size": 1618425 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh-events.jsonl": { + "sha256": "fc076f25f57458a5e5bcd86ec2e38590e862e520edb9aa3333701351c1f6c67b", + "size": 1824437 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh-result.json": { + "sha256": "3456629ae2c7c1ae75c6342cbc4dec842447ee10dd8f86320d5a9bcbc2e42eb1", + "size": 1401 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/config.json": { + "sha256": "99d3a7ba0eebd6fc74aec51effd7b245576915b2a727e6a021e0a83d0242b680", + "size": 1508 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/lock.json": { + "sha256": "cd7c44eb8fc0e15c745ad10e109b5beaed40400d010f4eeec0825d55e7d27595", + "size": 2093 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/result.json": { + "sha256": "6594d6b20c239a591ea5a63934e301bdb5faa48d136a9e670540bcb984527e8a", + "size": 38411 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/trial.log": { + "sha256": "bccea684dbc0cba80761124cd2cf6f4fab0a9241431e3df07e33c29da3f50907", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/attempts/1/ctrf.json": { + "sha256": "f69ddc358ac419c6aa8bb5243d169e501fe2a5b008c776224ed8cb9554ecf95d", + "size": 743 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/attempts/1/test-stdout.txt": { + "sha256": "344bf45577a2c7715621a2f3e7a669dce177019580c4a2a70c723f7cda4682ce", + "size": 1311 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/ctrf.json": { + "sha256": "f69ddc358ac419c6aa8bb5243d169e501fe2a5b008c776224ed8cb9554ecf95d", + "size": 743 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/retry-summary.json": { + "sha256": "62e2ae1d81da0eeaf24315c1b8c5de1a47b0d3dc17e0fd5cb84fff6e75fa10aa", + "size": 271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/test-stdout.txt": { + "sha256": "344bf45577a2c7715621a2f3e7a669dce177019580c4a2a70c723f7cda4682ce", + "size": 1311 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/job.log": { + "sha256": "bccea684dbc0cba80761124cd2cf6f4fab0a9241431e3df07e33c29da3f50907", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/lock.json": { + "sha256": "569fa5a10deef030afff7e11af56b3c94ccf90ae2594860463653b36b2bb85de", + "size": 3247 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/result.json": { + "sha256": "05a86be638efc7497d0c1a40cd7c72f1f6fda88f2a8e923026c4f616ea30739a", + "size": 1136 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/config.json": { + "sha256": "af06e540459232eded056d2429ff9bea502f8b94ff9667fdf144b93b6bcf8407", + "size": 3088 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/harbor.log": { + "sha256": "06790c4c6d068d11b994a516b6f7f8bf9bea1e6a131a189891a8c0308ddaa7aa", + "size": 1259 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/manifest.json": { + "sha256": "446398349da934e87a61312afab8f6acfcc311fa8a26264de0eb964c0be28443", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "d87b9d248fc7a7d544280edf2d80386361d2f4925b24d1bd7d577bc0f0cdc7cd", + "size": 581 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "9301889c4d1e12f80e22da3987b6798ad05059b709e99147605ff24c5032ebf8", + "size": 18700 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "0306556cb51ac05bcedaf81e75b7658088bd257434c79372776aafc493b9290f", + "size": 12008 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "d5b305e675c1faf78f6cd765a6a19fa1ca860887563dd5fbb5b82af743d14879", + "size": 15042 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "69bd709901db0daa8e22eb87a14ca7d19195e50767a47e7d206afe0717fb96ae", + "size": 18391 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "19fd20302d6a8709d30141f5d0ce2cec1471d4924cc6c78fc40ef3faf02cff08", + "size": 14690 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/summary.json": { + "sha256": "85892ea2193560ae34930533f88cc70405d8443830755239742a944b893a7f1d", + "size": 26712 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/config.json": { + "sha256": "805113349575c4f87e06c5676239746885dd5ef72fa299badd9277b452b7e1fa", + "size": 1564 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/bootstrap.json": { + "sha256": "a243941e9357e36a407ff4cfd4cf32498b67756cf083283184e3ac20269ed125", + "size": 2941 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/.anonymous-user-id": { + "sha256": "211a01de8e65d43e22f0dc989c794ad0b9a914bc3df851f93cc14618434ac8d0", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "7138119dab8dc22af2751b4499c5568f3852a2b0988775c7ac3f62fa79bc3acc", + "size": 127100 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh-events.jsonl": { + "sha256": "9053d5ec1e3f44061030b500717b4565bf2bf7917932a3c28accda4b73b8dd27", + "size": 140549 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh-result.json": { + "sha256": "c5a1fc593584a2899f1ea430958092205f0e7a9765c7341e23f4a8eb93548a11", + "size": 554 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/config.json": { + "sha256": "40697650a4ebf55dd1ef11449033f6d0c4e437ae2d09d6ce769cadf8330fc19d", + "size": 1538 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/lock.json": { + "sha256": "faa4aaf9cb07a52cd9942139dfbc250c818bd5020ecc1c695163ef7ebf3045bd", + "size": 2103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/result.json": { + "sha256": "26045760b379e3ea79996937a90daf2e0ae44214cd0a8e3c0f0e2fc514ccc7ca", + "size": 9865 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/trial.log": { + "sha256": "8d3533fc830ed2c13382bd5e4cd7159dd619909487823614dd71811f3c5c0baa", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/attempts/1/ctrf.json": { + "sha256": "bf350ef55b2ce37792226e27d4ec97cc107e6bc9f791ed2dc5c9b43ac1a74ea6", + "size": 2493 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/attempts/1/original-repo-ctrf.json": { + "sha256": "21246c902251289b0e428cc6bd46ce4b79284d90f0868f1047922c8e408cc44e", + "size": 136487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/attempts/1/test-stdout.txt": { + "sha256": "bde9f138d8ab9eb7a0e534edd3f4e2a68ce7ec2916e1f32efcc61cbcf6867953", + "size": 28551 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/ctrf.json": { + "sha256": "bf350ef55b2ce37792226e27d4ec97cc107e6bc9f791ed2dc5c9b43ac1a74ea6", + "size": 2493 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/original-repo-ctrf.json": { + "sha256": "21246c902251289b0e428cc6bd46ce4b79284d90f0868f1047922c8e408cc44e", + "size": 136487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/retry-summary.json": { + "sha256": "6e307bbd5557dc4dc6f55f6b7aa719ee94adc1292a1471970de1e41a14332cf3", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/test-stdout.txt": { + "sha256": "bde9f138d8ab9eb7a0e534edd3f4e2a68ce7ec2916e1f32efcc61cbcf6867953", + "size": 28551 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/job.log": { + "sha256": "8d3533fc830ed2c13382bd5e4cd7159dd619909487823614dd71811f3c5c0baa", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/lock.json": { + "sha256": "2f181a936b157c1d0b03280c36ca705ee0b5720978477205f6ce7ba7aaee25d3", + "size": 3257 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/result.json": { + "sha256": "7be854a6630fcaf4e536aa7c525db0bd878264a9bc0021f9899f5facab144d87", + "size": 1141 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/config.json": { + "sha256": "95fb2f6329d9f71e31ae9ddb1cc889e873d8e466dfc2043581afc2d22f04be07", + "size": 3108 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/harbor.log": { + "sha256": "a7a6185c50723cd69988a9307cba0bd7179080da7b03b521ccbfc4a22b9cc0f0", + "size": 1279 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/manifest.json": { + "sha256": "1b9a6fcc70ee8a059b5cfb1d9a1af34a34798426c27264c25924c43ceb8b1031", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "32e0c59bac6946dc394003037e768fcda289fa3ec201ee7e815a6076e9471fb1", + "size": 591 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "aaa4a1a8fe7b17f30131125288e10768f099ea75c7ea898a49021348155eddb0", + "size": 18710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "758612fe78f0844db288a7c4e8d05d928a686399fdc42daa431a8d17d5b5cf1c", + "size": 12018 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "be5bae52a8680a5d634f2c1f15b479c79d6ce07c9da7e102d6a1a6cf31e5554b", + "size": 15052 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "de1dca29406287e89c76ae7e9d090ee59c6715b13636d223c72b038a929a51f8", + "size": 18401 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "f3b47e927f001c480e5fe3f0e241efc4a47e1cc7afd0aa97e3b338e527a21ecb", + "size": 14700 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/summary.json": { + "sha256": "c819456f9a942f7a37ac7669fc3b3d517c0b7f52bd5378f1c431232b274c4857", + "size": 4044 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/config.json": { + "sha256": "c8fd6c668398c76d65c677ed202a2fd2f3a2b4087d349c364770ddd8c9f71762", + "size": 1550 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/bootstrap.json": { + "sha256": "139a6fd0776dee86f5a0306ebdadb3f1a4e79938a27aa2a5edcdd96dfa9be907", + "size": 4819 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/.anonymous-user-id": { + "sha256": "ebeaa0e0feec9b24bb32480e3b7eed14df490a2bc326a1add574c22c251df1d2", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "56d420f95c3c5e43d5fc5244332235865dc08f02429ea229d4080971d293ae94", + "size": 220031 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh-events.jsonl": { + "sha256": "4ad88ed8bac1c2b85bf361bfa47a1d7e5b8465aa7396f27246b2f637b091949c", + "size": 248446 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh-result.json": { + "sha256": "bfd9bb6b37e12d8ae4db50a74f696a1064756c003400e07273f11876bf873573", + "size": 682 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/config.json": { + "sha256": "1a60dcfe00ded28094103fea27fdd6a72f8abc54b183e4f325d3386e7c1ad237", + "size": 1517 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/lock.json": { + "sha256": "2d284ce2f6e6edc4fc33de774817dfb738cb957a8aa72516d694910a3daec1ec", + "size": 2096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/result.json": { + "sha256": "ebb44af0bea23aeca0f330db173e94063d5c2a0c7aac04df51ea5a01f7a2d91e", + "size": 12429 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/trial.log": { + "sha256": "1bc52dc0285fe52c4261113a7ca99e0169bf6db96776eaecf6eda89fe465ff62", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/attempts/1/ctrf.json": { + "sha256": "e2f2688ba84b4864795ee9586a5354334fd7b50355d69a1ed2b5611e186c291a", + "size": 757 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/attempts/1/test-stdout.txt": { + "sha256": "143fb90bbccce960c0dd95dad060fc95ac6ff18b55860c324222cfc86f62dfb6", + "size": 8913 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/ctrf.json": { + "sha256": "e2f2688ba84b4864795ee9586a5354334fd7b50355d69a1ed2b5611e186c291a", + "size": 757 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/retry-summary.json": { + "sha256": "30af07961a220d1d67ea80695c193564e627a58f47e7fde3fe24dff059c137cb", + "size": 268 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/test-stdout.txt": { + "sha256": "143fb90bbccce960c0dd95dad060fc95ac6ff18b55860c324222cfc86f62dfb6", + "size": 8913 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/job.log": { + "sha256": "1bc52dc0285fe52c4261113a7ca99e0169bf6db96776eaecf6eda89fe465ff62", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/lock.json": { + "sha256": "ff955393e6680c2ee280ef4f3864a4c5c0a2e714f92482d2a9280bb57f0f8479", + "size": 3250 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/result.json": { + "sha256": "309349835bc5d92574d306180c191d16fa36c78d0bcf560895a58e2606994221", + "size": 1136 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/config.json": { + "sha256": "f5663bea26f53ae2d68148781b108897e80139f774627c332f9429f84d776a08", + "size": 3094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/harbor.log": { + "sha256": "b28d840c89b8f31fa45f508c7d12a987bd04edb14fdd08a9029619b22f0ff1de", + "size": 1265 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/manifest.json": { + "sha256": "2f30ce0ec6c9ed8b8bd07f8bbaa530c5ef13746d58639b7641d9084d318e916f", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "c9d1aa833f18d51bc2d7968c6b7a59e2b665cf9f5178cdfb649c6b336fb7e4df", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "c1070e8b1e2f63ec27a251e067f33bacb19b797d14aae7be87a4ea5b0d2ea3ec", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "52d2bd070fc9fbcfde0634baa6d1950242da25bab37b0de7d3cffdf9b0e2a509", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "2526fafc0a69cc04128b654df1b5bd4e9d559089e99dd0f5af85292434d7cace", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "64f49b7b46d05b5494c4dbd4be3bf56149ac93a19e0d9b27abf7c8b03591614d", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "7cdcefc58b78c0c6b1e14f84b502c14d952b8603d97817d72157b6f3f2946019", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/summary.json": { + "sha256": "56f8fc47045fee04bb0cb9004441b01ef10790bada971eeb74031a31a0e74274", + "size": 6117 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/config.json": { + "sha256": "95d1b44729d32f175160f2929a8e6b7868f01c22e4636f317a88e9887be63446", + "size": 1590 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/lock.json": { + "sha256": "f801409b6ccfcc4307ac25c8d27daf548a904b3f407c825c3f518e12d9c9f329", + "size": 2105 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/result.json": { + "sha256": "9543b465485a93422a68d7741ee610603545d2e9b41cec779106bdef39c4d225", + "size": 4548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/trial.log": { + "sha256": "205cc24889444e76bcd3f8a76735b36de2dbe9a98b5551940c28428f242f2b09", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/agent/bootstrap.json": { + "sha256": "0c82eb159d150060290eebcfbe93607927cd4537bc16959451b2dcc648d61767", + "size": 6202 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/config.json": { + "sha256": "4c2683be6a7ec668a6907eb793428c000e3ee66c0288b45bcadd0e9059b22273", + "size": 1570 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/lock.json": { + "sha256": "41f474a2383b8b8a2438839d548493c9ad80292388f7b43d87b59d5942eca7c5", + "size": 2095 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/result.json": { + "sha256": "cc060ca2421d5d182bc554c9f5f9f003521ec0f4a25b9fc3fbce0bc6e0edc8f7", + "size": 4488 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/trial.log": { + "sha256": "fed581f4f9d9da9005d9c618bf03bf631e3f3817daa9f0933e00bc22251179f8", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/config.json": { + "sha256": "9b9fc76fec4adbcb5c083ab43557bdc091d843a99913c3085c56620f2d15a73b", + "size": 1781 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/job.log": { + "sha256": "b1b599d71bec8cab68bd9ec84b16cd7b4bff55066248a653c4bad0905619d798", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/lock.json": { + "sha256": "f373e1766626635d03689915f220455e6acc55b2c2bde8697b2b23672388155f", + "size": 5788 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/result.json": { + "sha256": "56906c181e53168918b0bc27cd02f04cbcd7615f6ee7699173741d4d3ce43b91", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/config.json": { + "sha256": "e7bbd0a3bebf5cac2bdd3c2857ef2c71c1ccb46226917b7cbdb917bbac4ea327", + "size": 3401 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/harbor.log": { + "sha256": "fd4848e8d69c58d45a6079396d6dd18da29c8cff245466c80d537708d68afd6c", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/manifest.json": { + "sha256": "a47999c0d83ce04d3b6ace156f7e73b9763cd4e495e9c51708547b99285a07ca", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "1c0e346ca862be6ab68aae0cf51db51884a3f2b2fe61d1343c70bcf93bc4f2f4", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "50271e2acdffb5df06bd14acbf1dd27c70e314e7ff7739b864cee4c80f80dcbb", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "8e972e614d2dcb56e7d46f08f78a468482dc984c8aeb56fa91c2e0226105e148", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "02759f987f69192fa2507ec499baccfc673ff543d12542b9d105cf2510bc98ed", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "24d5a0fc9e2b942833fc618003f671f47dd527c43de4181f6999074dd953f39c", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "47c015a729564997ee0f10a5fdb123ce332adbf7f9071b5311802ee6cc090656", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/summary.json": { + "sha256": "61bae69a5c87321a32a782a96fcd43e627d9c45049a7cf9cb24648e5d1143a31", + "size": 33635 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/config.json": { + "sha256": "84a69e4573bbeff1879f8630b6abf97a6c1a151e8027f21a40f3c49e217e890b", + "size": 1574 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/lock.json": { + "sha256": "b3dc477441c4010c78574e81dbccc827fdbdbc1346fbfb274571fd6037fa2b23", + "size": 2097 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/result.json": { + "sha256": "880f693138aae3abba7bfdb0a178aaadbc8c9629e329656d4ccda8ee84a93d1e", + "size": 4500 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/trial.log": { + "sha256": "044a707a488e221f09efac482e1d13ed9903771dc397e61c155168b6737a3cc5", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/agent/bootstrap.json": { + "sha256": "d75e36e5346e74a3f1b464b24b6e3bbf159103f626e474d7833ffed5bb202221", + "size": 15722 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/config.json": { + "sha256": "894e346fa69557428211eef86591d5dc1da9d5797d1ba2e097069c94cf8f14d6", + "size": 1576 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/lock.json": { + "sha256": "8799f5bfeccef4e613d4358b17da7062fd04e4d4b101f897804f980b72d59cfa", + "size": 2098 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/result.json": { + "sha256": "92b26a3a93ff64857ecd4b6d65b6853308aefb3eb3f190614967d1a790f3a567", + "size": 4506 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/trial.log": { + "sha256": "e476126d48d1b1515a872437f92dc9115ca87d08744b653c5a23174bd0614046", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/config.json": { + "sha256": "2f94c608393eedd3b6ec9b300268955740cf2c68503e4f635095b672c032a695", + "size": 1776 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/job.log": { + "sha256": "2307969b73d90e5ed136fb9e15059c7093066811795aaf79bc7ed0b53847d876", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/lock.json": { + "sha256": "c72889ddb25e9d90d48275624b4913349800d318b289233a622aa4847a70d23c", + "size": 5783 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/result.json": { + "sha256": "3bd541d0f8e586fb6d0aa985572871eed4333620f3418ba7d350f055e2382239", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/config.json": { + "sha256": "a5de6f01fa212bd300463bc1e0860afebb89da82f0102e59ea23c2d4ac9c1c17", + "size": 3396 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/harbor.log": { + "sha256": "9f50a7d42b9790f4af0ac8116a328d029a45edcef6de81fb7e9c02370c01e7f2", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/manifest.json": { + "sha256": "0a7a13cc673070c646be8d06fd8fbd2013c06f32a1fff6131d03d1c77755f95c", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "245144e2d85aefa675b7268997a0ce5aa9fcf2708b3a9877da62476bd987ca7c", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "db23509ff9d45bb4f568e70f8f9060119b86c77a447d06c18d34344990d3b39f", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "6de5efb518342ed063352abbe1b9a65cf3acdbf10fbbbc352be6b7c3d0335a6e", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "80f9f81d5cb42dc029cca977c83eae0d08c2acd9b2bc03c6e5b3c31215dc2b5a", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "43a556da597663df0ace1d2d0ef8f01aa3254337fd97e186902ab07f031906ff", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "57aeca2c1b66d2742d6b726d167eed9330488db2d7120027833a2a965c79f9fa", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/summary.json": { + "sha256": "1d57bea97ec7cd5d3cc348e065195ac5579e7d1a3786a177b8b3c5a0e2389a42", + "size": 42473 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/config.json": { + "sha256": "8a10abe4351b8db2d6d0f29c30eeae97da39b11d7bbbbf78edcb28ae821a491e", + "size": 1776 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/agent/bootstrap.json": { + "sha256": "317eb4d1a905a9446f48890f4be51b9acf3c0e230c0770b360285bae70d579b1", + "size": 12040 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/config.json": { + "sha256": "c929cd8b868077e73983dc4ae26f888af2ccd41f55e578d0fca9d2a3a6a990ec", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/lock.json": { + "sha256": "064efedc9f25ffabac629522a653a795ec7f84f17e5aed1e7fa4b4720e2e351a", + "size": 2094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/result.json": { + "sha256": "625c9a6e005e000984e7fc8111f481fd86ee259893d3c8f5cffe18b4b6291160", + "size": 4482 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/trial.log": { + "sha256": "a3d7584d63dbdeaffb22d28cbe5cef84c56d467ae96ef6ad47163d1c43783db5", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/agent/bootstrap.json": { + "sha256": "a243941e9357e36a407ff4cfd4cf32498b67756cf083283184e3ac20269ed125", + "size": 2941 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/config.json": { + "sha256": "140a73773802ca1ba3c20cb4d75bdbc4f6c53dacdc48679ab14c7b7cb4486d65", + "size": 1582 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/lock.json": { + "sha256": "50670f9989e3af3a96d3301bc92c30c1c987245ee8235457d6f6a89c04bc2bc0", + "size": 2101 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/result.json": { + "sha256": "26f8e8a79707a4f334422558031a0e2faafef53aad085ca04a688f8a2623a926", + "size": 4524 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/trial.log": { + "sha256": "40a6d92029c29f24a73990246125ce89f370ecf62916fa36550cbdfff7261818", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/job.log": { + "sha256": "3e1d01790c787a68e372427b850072d510ad582558685e5bff32cfc6cfcd9ae0", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/lock.json": { + "sha256": "f9d97b4b7f2372b23038b9f56e8a00170dfd17e599f19abea0b7b5197954d397", + "size": 5783 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/result.json": { + "sha256": "76d5af81fda0049bed001fde07d96ec13fb4c580c03283374b831c2f594528d3", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/config.json": { + "sha256": "555b3184b6294f6c92150b1eb6b37f4154c8925936db43422f5475de9da7d5fd", + "size": 3396 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/harbor.log": { + "sha256": "ddbcc011d61db3c49d3144532ff2451b744b4c5f63e4a7e51fa9d7fbbd325f08", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/manifest.json": { + "sha256": "2f35cbc799352d952caabace8f9ed3eb9f9d6bea62f390088e9c92698362d61e", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "da6bfe5d4bb9b1f132f305071d7870a4b3b808d5d4e6ccd6e45def03131150cb", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "4b1c2e567d82241189ac458a7a135239902d6d006a37b5fcfabfa4688340b9a6", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "9b4d8a2920633df2febdf9c3d28a7440c5734cb021533ffca096ab648b5780c0", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "f0fb82b46bf68fb0600224deadf6a19d68f666f665b2c99103e871fe1c88d8d5", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "3aa76d0c8fcf7a3696cb04d5a0bb4d1fa0d11a9f992097edb844a349b9181317", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "3ade0a7c4a9b854468a01f77bb6d6952ba34386cc0f6d6c35b51e51d251d15b3", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/summary.json": { + "sha256": "90943816f7581a3676aa4a7c4f6052ec055de61c1de145e8c58c17af03fc400f", + "size": 17449 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/config.json": { + "sha256": "bd986d61c52105019ba39a89f1097b7c340daffefc313942251a76c7775e1dff", + "size": 1767 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/agent/bootstrap.json": { + "sha256": "139a6fd0776dee86f5a0306ebdadb3f1a4e79938a27aa2a5edcdd96dfa9be907", + "size": 4819 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/config.json": { + "sha256": "89bf9f0ec56eca85c92aac898cf5ce6725c10c639f7ad55bd7f8592de4b0eaaf", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/lock.json": { + "sha256": "a1a55723eade55a9be2af6de257e6ef99c316732fd9d404cfb58667327c2e9e2", + "size": 2094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/result.json": { + "sha256": "aa8f47a7aa4d3e558ef34fb2ee22f9ec7cdcf7c44f4eb0a54c64e5d204384ac5", + "size": 4482 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/trial.log": { + "sha256": "6d00d11376937093cfe030b1c1bd954f7c55aab47381857ef636dbce270a8aca", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/job.log": { + "sha256": "ac91d10c1ec68dd4514d8cfdab010e4a0f1ac92d950d3ebebcb17b926c370955", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/config.json": { + "sha256": "9f488cb90de8f34f469054df9bdaa5b6f7e3c956a675a21f860e13c15d801f93", + "size": 1564 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/lock.json": { + "sha256": "23e256c941bf5f7dec9b1ab1e28c361d38fc43d12eb6560415da144c5641f5cc", + "size": 2092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/result.json": { + "sha256": "ed6e15f3638c215e50b21a7dce5a86eebcbbb07ad5778fa02eace1d89371baad", + "size": 4470 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/trial.log": { + "sha256": "0d6fa24433f9afcdfc6c4b8abcc8b62c312094a96724c363cbc6ae813c7ffbb9", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/lock.json": { + "sha256": "8168e3f78f90873681e470b477a382b30c536fbe59055a70a3538024168c02af", + "size": 5774 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/result.json": { + "sha256": "7cfd1e8a93d9b4ffdb641fc00140a85d8a6d28c3568d304a8028b10edb593c55", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/config.json": { + "sha256": "1ec3767d598bb0f77efe9deb23d34c85fd4017f800836975b59ed78c8e425777", + "size": 3387 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/harbor.log": { + "sha256": "f51ece1a489bf69f2f8ef7aaf8f0a51313ee4c630b27781a0b2ee582954acc86", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/manifest.json": { + "sha256": "e46c64ccb95f0ed5b47e11db1c4404df71ab21a2096ddfc3df560a3e3c8a36d1", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "59392aa346b738182011648d1c9f82d7b6e8e427a92bcf5335910f6a2e0da073", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "a1fbfd1178a357cf3f9973af0e3b6d9e5d5863690821982c198dfd12beed134c", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "3baf0c9bca5a31209892e642ef48b4b8a6c97ea53affa993db52a6b9330b54c2", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "c5d007d0b61acb7be366857f4d7edb79ce22dc77903d07e12e484388992e2be0", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "d68c9722bc55e152d50f30875ed97bbb2f2ac8f59afc2ee1bfed1063a003b725", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "73b9b98238e2c9cc7bfa462bb9d8b2febd380a399670e4155d72f8e6615c6b5a", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/summary.json": { + "sha256": "1262258e1c174e5c3e0cae701841e515c4e61714297ad141f6ba5d7ce637ce4a", + "size": 30250 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/config.json": { + "sha256": "487897699e36c40ea684fe22f5d72b864c391d38933784289e3fbd349ca00e7f", + "size": 1785 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/job.log": { + "sha256": "41a319f6a5a5c1f3db50e7f287aa35f99bced1929cbfe69e40521f034c6aa3ab", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/lock.json": { + "sha256": "2dcd5302dc8fdd3a8299cdb64b0cce8188b64c20b147bd6640aa81aec203a70d", + "size": 5792 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/config.json": { + "sha256": "2997a458852ac5968f2f759f780942caba75161b9c5e253833218d9b6babb6f5", + "size": 1584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/lock.json": { + "sha256": "20c558585c80a0fffc5a757aeae23463465785e2e67840240dda2d0c7bcc7c42", + "size": 2102 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/result.json": { + "sha256": "2c5d6232aeec9ee5b4c040a0a9ef0e5e76f360bf880579cba4bae9d10b69ad5d", + "size": 4530 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/trial.log": { + "sha256": "58121c96b76e71cf587af2c270c339744cf12879ed84866ffd464fafaa87806c", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/config.json": { + "sha256": "a03bb3ca49abc6cbc59723f5e9a1fefd96ae5e7ff668be7f1968f344f0e0ec76", + "size": 1584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/lock.json": { + "sha256": "d025072f12d2b1f2b215632c1e7fbfd3bb736cb8b0cded1afa75beb54c1683f6", + "size": 2102 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/result.json": { + "sha256": "2cdf1b3a5485cdbc6db2f390c5186dec2939d374c476543f04edad59a0c66a14", + "size": 4530 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/trial.log": { + "sha256": "31463526d03b19f12abb58479e73d9c6d7596d3511f69154e9545ceeab919ccf", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/result.json": { + "sha256": "f4ea78d4cbe3c2bb8d7d7aabfc1bd59553e7550ba4e310932819b1c22323ee2f", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/config.json": { + "sha256": "cabb182449c5eb40ba972e754be4ba079b3031f8f5e6ab5555a85b8c60668e0d", + "size": 3405 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/harbor.log": { + "sha256": "6b32941aebc6a05f335e779edfe75013953ac434dc696ff559e3b925fe94c293", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/manifest.json": { + "sha256": "09adca648d3cd50c864e95c0b3d36326b7db09b5227435580427cccd893102bd", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "a281ae62cac4ef1b3811a42e6f61c3afc004ce78e7b982bada06e5b55e8e446d", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "e539eaddc4b82cb79a9e5f6001c18d172e05c37903f2fdc187228afb530e4eda", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "aec184a3dafdbc921e2dbb2c789181b9f3254ae144bcdfa53f0adcac5ca587d5", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "ed90c26d1f01ae05adb185141fa30f097d3143d734fcf6f98bf2de4c30dd829d", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "e8ae62c404fcfdb0c862e282c0f4360641130cee3f80abd8c6f8156f32ce3a5f", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "c9ce6735d09bdf2d895232b0d061b46778559a89c426c0a21a0615095d9f214e", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/summary.json": { + "sha256": "0196f9f1298444e0a42ff5070ccd7f89717ba14cb8bc5edb34ec404e066de7b4", + "size": 50884 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/config.json": { + "sha256": "318c60839dade48f7c88495b27021790ed67df9c793261bfc28db356110f90c1", + "size": 1791 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/job.log": { + "sha256": "7fabb952f0ef21ba87bc372af05ba9ab1b1ef5295b8a49e84f244f8e6b1a24ac", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/lock.json": { + "sha256": "78567e4295eb199141aefcb8f529915ef220a28fbe3d743eb062a7fabd62d0c9", + "size": 5798 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/config.json": { + "sha256": "ee80135e7236b24d1701633e04b1a0eda4110633471730cfba52c340caa9bf91", + "size": 1594 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/lock.json": { + "sha256": "4b0afdc1d2ebec39d7262325410a63d76049bfe9bcf5cf94d4ad085f8840ee22", + "size": 2107 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/result.json": { + "sha256": "1cf09cef6d5f195c6f64e1dddf49fe52fe61583e8a6ffaf75bb9e6454ae2c99e", + "size": 4560 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/trial.log": { + "sha256": "9a56844c5c431c33f46aba516344761eeb4e1166f2c850dbdf83c284820401ed", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/agent/bootstrap.json": { + "sha256": "38a018af813827b21bff5685e056300a44dc7899ae803e02ae564568808be3fd", + "size": 19659 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/config.json": { + "sha256": "b621320273ca4636111c944d12de3861797412fb11ff725ede2ed3cb76e50158", + "size": 1586 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/lock.json": { + "sha256": "2c4a73214e32afcc0eb191316dfcca9ccbbc3ac62691d43caa3ad5b51f6c0f51", + "size": 2103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/result.json": { + "sha256": "cad316ea002c327d02d2eebdcb379f5cc2b631396cefc82cb34d306f4e92c16c", + "size": 4536 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/trial.log": { + "sha256": "5acf988764cdf2ac7228da80967df18b2a34bfc609ff246a8af0202724043409", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/result.json": { + "sha256": "ab404f098ad3f84b7704af1baad0c7cfff7ccce805ffa0094c607b751dd4e1cf", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/config.json": { + "sha256": "c821dd8ed6043fa1a552ada257eb8442f1978a08c7228623bdee78257af56755", + "size": 3411 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/harbor.log": { + "sha256": "ff8a4592b2ac0a707ea4ca3249d73f622c29a48979ceb694271c8aed6f929f53", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/manifest.json": { + "sha256": "c0f6916a703f0c9110e958bf0f797a2b4a261f65c6be3fc137d3a54daf01db73", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "db1f9dc07abd430b730355d9c35aef96e5373dad2cd118cf72169ab625c34d3c", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "118669a7e4b0b783bd0722862b80465cd94efc3df2db1f6690835741dfb77d65", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "3f43b0999974e9c16c102f64b1838d45f60a467b06c588edd645ceb88bece261", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "c8b83d70d0188bc8804eb391519659c822b8ae337d7e02fdcbb1715d0490d075", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "59bf4a22bef042042db83e0a5929bc0fc762f4eff444e25f3e1cb7ca675d91d2", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "0bfcb2eacb71f657bcd0028a1faedaeb661091fe862e4c261e5a22150a24a685", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/summary.json": { + "sha256": "b13d0543f0cb1c0f367e6f74317c2fe759e399297e2061665719c64847dfdb22", + "size": 46944 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/config.json": { + "sha256": "f0af1f6e8c1e34bcf8e251e2de0d19536ddf9add62c58ab4aa969a48bfec9e16", + "size": 1783 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/job.log": { + "sha256": "f486a6ca446bd9ffe36a178bca7322179038b3cb8243f3a8dd0f95e36593b9e1", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/lock.json": { + "sha256": "f532cbe7f06293e205740409ab3ae669fb5d6c7ac9fa0956d1d3ca7fa86cd7b6", + "size": 5790 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/agent/bootstrap.json": { + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e", + "size": 15490 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/config.json": { + "sha256": "0399631455909f6e9b9e25ade2aa4a06344efffc126933d48674078915ef63d1", + "size": 1580 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/lock.json": { + "sha256": "fe48b3fe222e5c9455315bb25dcc5bac459491b2be28fe55ea790bbc8f43349d", + "size": 2100 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/result.json": { + "sha256": "5f06c592c06135e40b91172af316333c4edca880235a1f386784ea97acc24d59", + "size": 4518 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/trial.log": { + "sha256": "48f7a2e7b62024ed77d682fa111ccb6aee32143797a0a030e81d9290ec156add", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/config.json": { + "sha256": "c4dd3684b9ac3be6348068820a8618288a43629adf488f82de73ee25babe1ae8", + "size": 1584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/lock.json": { + "sha256": "7a7fe5f350aaa9b7a494dd59d820ec2a5123f7b233c48ddf9ddb2e98e9919c14", + "size": 2102 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/result.json": { + "sha256": "6b07e89c5e841b478bfc4241ecf3808dd391f9cff3e3e95dff16d50a6d28823b", + "size": 4530 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/trial.log": { + "sha256": "9d3c8093f68087c09eceb542bf485194104eebc834de8455b97ee4c764261309", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/result.json": { + "sha256": "fa4b33519eb307fc42a52b1482fb02d86e7c541446662e2ba29d3b1fcbc025ff", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/config.json": { + "sha256": "9bfb491ebaf0ad8ce23e719e9a30caf702f1b0b8e77aa3e41821a5111794b723", + "size": 3403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/harbor.log": { + "sha256": "c0c97bf26be8ff81670e111e0c86c2c615e5d6acc7a7791345f0d65a5cadd2a9", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/manifest.json": { + "sha256": "59297ca64e29863872cfa97ab882a77328d38ad7f5584a6890489c8825f81800", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "f635e57928588f02579d3a4a9e36416a644a08bc69d68dde1382ac05008cdf9a", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "4058ee68ac43833b4110ec03f452a6179ce0fd3189fbe152337f9c11a566c3be", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "c1230b6d8fd36c58feef9fb31511ce19d1b4290a2940b752a95e66e5a2fdd4db", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "f428c1abc65906ddb7303d4038f7c899c831eccaa21dc171f5c6209aef763461", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "a9e7d10ae95f84e32b18dcbdb68e10c1b8034c7d24dc68ab624d064d1ee3f753", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "a8b715a848264f73de40100f05f78e5fe1ff8b715a0c2b5dc883e763c381ede1", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/summary.json": { + "sha256": "37b07dd932adf6c214a49be3fcc88a67ee069f1d178d036b372829d6ffe85a48", + "size": 42255 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/config.json": { + "sha256": "ad97653639ee1e3cb0e35d74faa16a5379a8153ca58b17d3d830dacdb1c0d0d6", + "size": 1767 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/job.log": { + "sha256": "156662dc197fd71892451edfaa5943045b93b24895a063d7f5abc6cda87991e2", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/lock.json": { + "sha256": "fffb2eb15f08231d626ff8e42f331fc0522ddbdcfe1b8f336e6728f36ec01a80", + "size": 5774 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/agent/bootstrap.json": { + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e", + "size": 15490 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/config.json": { + "sha256": "165e3453ebb5299ab338d221e99ff83a9637a6265d9c4484daa8cde47587e8ca", + "size": 1560 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/lock.json": { + "sha256": "55f9d34e9ba06d2619bd2aaed45616f3d26ce0c544f34946d5e8813d9b09e675", + "size": 2090 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/result.json": { + "sha256": "73709f5a967b2caf1082b5a45144fa483bfdb0643f599e792db0ad5da11f3a9c", + "size": 4458 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/trial.log": { + "sha256": "e27b1d195d1a13029d968c1ba86ac57974f0693ea003fa7324d4664c01d3bb96", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/config.json": { + "sha256": "827946adca8d9cf649d9a2d9474d8a9c7d5f9405bbe05f1eeb899755f3696a19", + "size": 1572 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/lock.json": { + "sha256": "7aab5dd83fdc9af4d7a0df7aeaf0d31324f3591bf51ca8e44baa142df2ec4e5a", + "size": 2096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/result.json": { + "sha256": "a50ce1fd10a899155353e312aa5ef356fb05194cfba6f73e899c9e9da4fec620", + "size": 4494 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/trial.log": { + "sha256": "81cf755275a66e3684df625a777c0a2fe92b05fe826ef80b9531ae8888842c62", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/result.json": { + "sha256": "465cac16129c63b3829019d612e323f6fcd89c1cba477166e0aa8314ecdae2f8", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/config.json": { + "sha256": "cb5dfa97cf0c17ece46c8261ed75faffb2c201cbda0bd1c7e09e5f6f0c31f1f7", + "size": 3387 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/harbor.log": { + "sha256": "7e959e4a77e248c25d9b1796b9080ff9d1d07ad9836d7de89624763431608aa0", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/manifest.json": { + "sha256": "97bdcff4c34faf3d787c3860d68f8b958812d6ae51f20ed00012b62abb4da721", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "cd5b1e4fcf857cd69f9dc3a47fba6814e2cbbba77afb91dcae7bfa00046d4b7d", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "4df5fab3f6aefd6ae27b14dc0882c3e63f611ae2d8b6a759aba1246c49263e85", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "b4f3ea586c0a7200a2012a7dadce18f9d83815a3b368ce9dea0df6e093b0ca0a", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "847e009cc2129fbb4de9978bd4bf1a61e52aa3e25cc2efeaccdafe238f15915d", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "170ad2882c900062e9b5d4d11aa3a6bf524ba83d837e61b26bffc138b0b3c9dd", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "37f9398851afe20cce62a72b1d73f69a7cc811c2b5777affeec60425ef8cdd56", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/summary.json": { + "sha256": "53288df2ea90779162c20bd6ebdfc7b38a8d38f90acf3794a9e47f2ad711e2a7", + "size": 42223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/config.json": { + "sha256": "d237c583a1bdae77adce87474bc5a8b19bef72c21cbe099a5995969d8928b47f", + "size": 1776 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/job.log": { + "sha256": "cce8916cf195193d1742824a4086c1894487e3b5e9484fd7fb7bb536f45de71c", + "size": 14932 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/lock.json": { + "sha256": "5ecdf020314a62f843f6e9b46ff85301cc853870455d1b565d6e5bede0799b26", + "size": 5783 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/agent/bootstrap.json": { + "sha256": "b97b124b61cfc3b413b29b69460372d0b964474f0b96133c4a5571d9df639c45", + "size": 21730 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/config.json": { + "sha256": "49ce99bc3591c7b85b3c811867850231702601e6934ffeafa29a579878e3f205", + "size": 1582 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/lock.json": { + "sha256": "82f06a9eb95c410cd4b02a41d7e7a0a8e68205498cc220fdbf23780306edc7ba", + "size": 2101 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/result.json": { + "sha256": "6a7bf4ba88b91f90a6cd845403006de53b7ac3ad2df1db84d8f09ca0125730ef", + "size": 4524 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/trial.log": { + "sha256": "39b28f3495a5578de7be930dbd13e3691d08bbc85c02b2eed4183226fb00c030", + "size": 7465 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/agent/bootstrap.json": { + "sha256": "1e672441c98d4e4e39bbb460a3c2c7e1d185954628be757adf5af4f0eef9a767", + "size": 8485 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/config.json": { + "sha256": "868a976eef688b7162bbf669d2ce0c1e4dfe21d526eb367c94b12b3ef3bdab4b", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/lock.json": { + "sha256": "34fe23312cb9a57668f373fee54b3396ae980e7bc2b73db0a3f2d49ddb67012a", + "size": 2094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/result.json": { + "sha256": "0e3ccd245745a31c01eadf8d10e4ae5a061d083fc07e828211a99560cf9196c3", + "size": 4482 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/trial.log": { + "sha256": "3e5e57d2cf1fd4988ef34939df67fd8fdd8fa871aa31bea56e2ce31df2867114", + "size": 7467 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/result.json": { + "sha256": "b7aaa14449b27a81e8d7e5cd463822619a76bd9065d51d8e934858cd85fc908a", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/config.json": { + "sha256": "5d5fa2ddbcfa323dea44ba60c1a8adcc2ecc566280366030b362a06ed2caec8e", + "size": 3396 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/harbor.log": { + "sha256": "1b772c3579e9c462009f09ffa3de7b2ddcdd28b444aecea896679d6a7669b99d", + "size": 1050 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/manifest.json": { + "sha256": "00a879f86f550d9aea4530e2fd058f04890540a26c0b2a15f44b2d5f50b01803", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "1d542e4784d0b193dc58939bccb156f0d16c30c3ced1f0d94dc9238b9a803e07", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "10d6422cdd2bafbb3ab71b9fc48d2fa95336937d0bd4e021e396e5fd8fd64751", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "b4d803b7af0b632eb815dc37f8a61d8cb76ff02ed282cbd7bb350096d0db6e01", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "258f2908e634e367ed124caa82fa5c735cd2ca97e1a50e74179e079c146ba550", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "1d7e9bc6a4881b953c1116207d2ded21959a080268072ccbb8248a20148cbc99", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "b0b0c58fa3e7e788dfaa3d0fc46fd427b04db3e2971246f418b121a64176da9d", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/summary.json": { + "sha256": "4f447719c1b8d78ab9aede92dddd9145abfbfb292efc75b40e7cffac7184ec45", + "size": 34603 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/config.json": { + "sha256": "0e9f0c233981ed96357197df090d35dabaf937a6ebcab24a82dbbd53059ebdd7", + "size": 1762 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/job.log": { + "sha256": "a0a9436c182ddfde3c6cb6f5c03f7f2f8d00910623d6fe5118610e419b3ba983", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/lock.json": { + "sha256": "2e45d443b26ca73f4d7b8b1068bd02802f15ab33876be667991322679a545962", + "size": 5769 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/agent/bootstrap.json": { + "sha256": "89006797609af114ee0903b557c5ff434b3ef6071d8346100829e653fceb38ca", + "size": 15103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/config.json": { + "sha256": "fee6ab6395ad7d4af9260bda2cb52e602b03e90148575b49be75d97ccbe7d6dd", + "size": 1566 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/lock.json": { + "sha256": "5551098d80812747691783f7303d186dec5249e4ba4e1c9ec4c59182964ef025", + "size": 2093 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/result.json": { + "sha256": "6fa73803640f1d1bc557ef8bac23af91149d7f55c10bb5c7112e45239115812c", + "size": 4476 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/trial.log": { + "sha256": "66e15bb6731231cf1d81c4daeba11dc4a288d2781e464d4ced1cf9b95856f5b4", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/config.json": { + "sha256": "63a8d5643ab3c9181be48c24705fb57070e4958a462c72a066e933809c993572", + "size": 1556 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/lock.json": { + "sha256": "bf369db4c5b364f363dd6105ef05ea0e11071c12ea810fba867ed166bf03b74c", + "size": 2088 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/result.json": { + "sha256": "091663f68b91509c2337d427b83e75b3b4a7df9bfcad6e94afd4ec565831af3f", + "size": 4446 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/trial.log": { + "sha256": "6c95b8bad75b2c82a7a30ac236a8e36e9037d10dcae072e7af79ed36dcc0c2ae", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/result.json": { + "sha256": "77f27aae6bfaa5d8b012f29219713ebc596abe170db3312212e4136920b2a1ec", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/config.json": { + "sha256": "b35de5420fe1bc3f9f9a9c7bb7ff6d1b351756d5ebe6fce26568040f4610cf84", + "size": 3382 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/harbor.log": { + "sha256": "9a6cce8fdd7bc3cca334a03c0a4871c051f6447cbc905b1f9ebcdc3b7ae9ee2e", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/manifest.json": { + "sha256": "502459c6482d86928087898c676b5390c2229d0009c952fb02da783352bcf51a", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "ca7a52c7be278a0ded6e1d0ad9787ebfe46d1ff2ac453e4d617e504eb1d5b9c9", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "129b55b279d0b46a993f4a991422353d867f8b10e66d8d28f6f5296d8c0455d4", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "6fa3f5f89c4615b822384d101c3dfad9932eaeec03ec6717b5b3049d7500ce84", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "0d2c75c6ddc6eddb232d52deb2691bd7e16f649227f53390e9a89dbfb411ea54", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "f454b00ad2bdc69c690482dd04d0d8052f98f9ab15a7126706ec5cff81000b58", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "9ce7f857fe315821440a8fe51e968e7ee5361ae794cb63a4a1dffe1ca31b7322", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/summary.json": { + "sha256": "d6654fb376206566eb2d98c82cbb9f0a3ecbe0047456c5e74878530ca465e258", + "size": 43548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/config.json": { + "sha256": "149ee7db8ee5da5d1d717688d97c75511e8b286f3b4e755ce3f2359e5b3fee60", + "size": 1789 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/job.log": { + "sha256": "2cf1f5389e529361d8edac8aa20a6c9a0e70a0e6f258d9e1a4f29f2357629284", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/lock.json": { + "sha256": "96a5e8573cf55b5ec6296c73ac0be16ef7dce35fde6e8dbf16ec113e4d3344d2", + "size": 5796 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/result.json": { + "sha256": "b50aa9ee9e9216b5c3f192242dc7cc19f6563b023cfea814461acd6bfaf52861", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/config.json": { + "sha256": "21fb871551a0205463815485bbeb97071fdefe4f93103a92601dab1f2d5f3eab", + "size": 1590 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/lock.json": { + "sha256": "0618206f3bf18cef9675c0efaa21eaff12bc28f72e7585988b153809d92afec3", + "size": 2105 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/result.json": { + "sha256": "bc184f6f88666ccdfe68e25811e45b5ebcacc321b247307bb88bff5408efd968", + "size": 4548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/trial.log": { + "sha256": "3ceaeda42f39e0d82d60737756070615707150599e50eca69dabb341b65dfa54", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/config.json": { + "sha256": "c5a52c6dcd9fd597168dccbfbce119e7f3800c1152e842c74cb8b4483e7b0d1d", + "size": 1586 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/lock.json": { + "sha256": "34198b7f2761530d80e133edde12f606515eff0e1b7f2d0459493e1265e39d9d", + "size": 2103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/result.json": { + "sha256": "5ee413a851303fa81a99928b7a8b2653bf933f2452672b1b52841cc827edaf7c", + "size": 4536 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/trial.log": { + "sha256": "75b6341e0b2ba64d0b6e0bf45b729c86974d6c8a408d7a85ef87e9e5cf0e195d", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/config.json": { + "sha256": "9e0b34d2fa8045e006c2befdbfdd724a708c115aafceaef0e1281124b5295182", + "size": 3409 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/harbor.log": { + "sha256": "602b7756416bc15295c34214ac406ab8298b3200015e63305f7b688d5426d3d7", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/manifest.json": { + "sha256": "8b148e8e8bc4495d1806a39c7e4e37d1690172ff925179890fd0918f4583365d", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "0527e54a4545a132ac8b23487b8e85f578122f19f5266090473ae9735414f033", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "8efb0b5a5c76b092af22210b7ad1c302711b466709335fb100e38c7e6055a9a5", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "37b0dd1aba7c948d6f0912a656aa8fcc3c752490fe87b5241d9c46492bc76c4c", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "15517634fb9b35b0e4d4076044fbc7fb1594babb599ea58a80efd0158e42437d", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "0200a7d5b6fe3c7dd4f9e5be30f513fc935398a695e85385b4b965441b7c090a", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "21b0b46e0f669ba0064ed87d7c02cb4921880cd0625933d1677c32b3d0c02800", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/summary.json": { + "sha256": "2014331be4e6536020bbba6ae8705958c391f21cf40b4cbf45d042eebad3a727", + "size": 52698 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/config.json": { + "sha256": "6bdf189458b7f0cc6cf21d58e683dc7bb7824aa6cfd801e1f091ce790afd01ee", + "size": 1777 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/agent/bootstrap.json": { + "sha256": "cf2881c5ea9a508fcf319b2a6befc60d39bd25004a7d4f3d2b194579b31fe401", + "size": 18592 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/config.json": { + "sha256": "48525a60a3ecd38b4fbc2821a495c092ac329c094dffb22f808ae74145bf8e82", + "size": 1582 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/lock.json": { + "sha256": "da3a72b6720559a990b81ef80b5a4600979a9f45f4b816c7f00aec760cf850e6", + "size": 2101 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/result.json": { + "sha256": "6aacea30f7ff30f3913b0b76892263bf23add9253d86e0f76ea0b8fe4b1067fd", + "size": 4524 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/trial.log": { + "sha256": "129706205dce6e784b0b3a88d6e13146794e264b6e2ef7769572ef37df67c27d", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/job.log": { + "sha256": "3f14319b8080fd75bea2313de83791aa02798c1e4d7a117e6df928af207e9632", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/lock.json": { + "sha256": "aef476f58e277726fd5d1b3ab6ed164c643f6e2bbe4bf20c1532b2ca514848ef", + "size": 5784 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/result.json": { + "sha256": "0a050e40fa06128800219be847c001a387b6de3fa1878ace68c83990fd4b8a2b", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/agent/bootstrap.json": { + "sha256": "106fca350e150fcc87b6003fc699eec6e057797103b83a4c169cf5c14a6b293c", + "size": 19374 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/config.json": { + "sha256": "aa8c987cadf7bd8c83055868fc9e97d857e054c70512b1a98a3bfc44ef53059f", + "size": 1570 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/lock.json": { + "sha256": "001cc94b045a49e01e652f660702cb68e227942f1f908a25a9556d8a7e9e1204", + "size": 2095 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/result.json": { + "sha256": "1e5a5194aedd3712d9af22c4a6c25124068a632a8e8846e7c7700820acada7a1", + "size": 4488 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/trial.log": { + "sha256": "d68a591ea6376bccc92f58d2508fa1afcf7a73f0b2d23fdf49b7cec7ce6b8aa7", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/config.json": { + "sha256": "2b49a82352915aec5ebb2c3d4a93e0deb0f2278daaa0d38b5dd7efc0c1da52a7", + "size": 3397 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/harbor.log": { + "sha256": "e70328d2a7405342e019275bc58d2c9348b6d7113cba5402ab3ac9b8fe268af9", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/manifest.json": { + "sha256": "93bcf89c9a1af9248e825cca85eb2bade9579b60a8a79fb12ff0ce7e8870537d", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "5ac1da16ac853c394cded71dee3c77b699717fe7a9f7e36ada6ad07c12739d91", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "9e84f9fbfa606c1f7a3599cfa79fe9fab287532a7cb72dd064ffa5b322b3fb37", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "b152bc7bcd234a8d9b8591fff58d49656ac031e999c6c1babeec52651d4713b4", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "aa15c9df9bacceb9087cfec15b7c0e5819ba67529a198a1cc19a24b32e021b35", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "0ba69f1c9fc1b004e6a5bb89041dc4cad4fe711f83b922e70443d777e4b94ec0", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "1cdab91ead3231dc44b97101ffda0374e13d85cd1fe9a0d3ba7ab1fed82d18b6", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/summary.json": { + "sha256": "6f527aa51e0909c36993a35e0784ed31c01418d97d8355d745e979cda8351577", + "size": 43082 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/config.json": { + "sha256": "f414df01990b4e8e24c107ae0bbb5541f55d34609f2a2d15c6d7e55d4bf8a1e1", + "size": 1775 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/agent/bootstrap.json": { + "sha256": "747599891fefaac25e7ab1119fee62bb88a9bf9b3e036a496c37c79d15610b74", + "size": 4819 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/config.json": { + "sha256": "58c483f1d9461d58a96f37bdccde3efb6718e16811612c361a8838621ab9fd75", + "size": 1586 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/lock.json": { + "sha256": "03da824fdfd541df297fec10101ac25d343b300202d0b4ae235a1228927686d7", + "size": 2103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/result.json": { + "sha256": "8e1e193dd8d4b1bfeb8cb65a45a488e946deabf848ea8322e28bc630c0465449", + "size": 4536 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/trial.log": { + "sha256": "0e62fb95e279def3d90c2f7e642f3b2a12888f3b9465d5c3a76cb0bc99dfb87e", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/config.json": { + "sha256": "c6e2a01f2055b867b528c0c8655365419ec6b1d4427bc3101d4a155f921922ef", + "size": 1562 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/lock.json": { + "sha256": "2b5b37e3c568c708c3208e51c5c8843e6810bb319728d6a6e24297d97aa12267", + "size": 2091 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/result.json": { + "sha256": "b005dd171d4df9c4d790f4518fad8122370217d3252fa14b9bd6cd4c21508aa7", + "size": 4464 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/trial.log": { + "sha256": "51dafe92cc686077eaad952a7b0e8c062c61aefbf37b00a9b911a3bdd3fed7ed", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/job.log": { + "sha256": "3699b214fd44ccb9f13f42c744102201186c9814a3ea10c21cc12ecb81f2dd8e", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/lock.json": { + "sha256": "bf55d4d6e307b0c0b2b75431f28d360af02b35bae3896609cdaed4fd50f1d2e7", + "size": 5782 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/result.json": { + "sha256": "7020e12bd369603e1f01160edff28b3822f65c12db630103ffcada643598c56c", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/config.json": { + "sha256": "3a22ceb7886ccedeba76b4f66ef290d73b226b6d56ff420ccec98a4a138596c3", + "size": 3395 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/harbor.log": { + "sha256": "3152bf69268f5933e92a8da6c216edd8effa738d9cb1bfa8707c26e6f74b7bcb", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/manifest.json": { + "sha256": "eb140223825530b724c6b4a7455a51a44b4a6d9a0c0048a20bab069bf7111446", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "73618c96fc5c9ac1874e873496d00a89c064fb674868f85c84e8017adaa8daaf", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "f5bab33280f5c0158b79adaffa7c5fcf748008dcefe53129ce3d0088b10bcd21", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "d1f7a4620a3853e230754db07808cd9438b844e524cfc6efa83fc8ab9c57fde6", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "6c50357f287b025fdf5b6902ef81c883e4be38c2c6f90631918179b2403491cb", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "f03a1e7e0c2f7ee018becd904e5e100b4e221d21e33044715267a9bf9724a3c6", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "641ac77a7e5643a6df85a67b5b7df9a188daaacaf739b8687badf208c66f4360", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/summary.json": { + "sha256": "995297252ae987cf60c3d4b780ab768e4b51f1bf0bdb0c2327fddcf92ba9547c", + "size": 32072 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/config.json": { + "sha256": "ba9b08939a0a1e47ef24ec7c51e29d30cf402fe6d730acf6f5452cc1f3ff802e", + "size": 1783 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/job.log": { + "sha256": "9ca42804ae324a00201fdfa7717b1cc5a679bb553cb1f419078ac68ed5927a11", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/config.json": { + "sha256": "dbeab3eb020e592c03f23829b3bfb13b1fc8f25b49b91d7639625c4358805309", + "size": 1602 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/lock.json": { + "sha256": "e1518d39192a79ca527776d87edf6c95f0a3118b6058bc3322dfe240d3ac1011", + "size": 2111 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/result.json": { + "sha256": "d9d86f9778f07453b8949b1eafe18f8349d059f247d623410e90f63c9a504561", + "size": 4584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/trial.log": { + "sha256": "918be509525116b51289b0fffc2f088c11665e1bb3e540a6b05768057484d593", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/lock.json": { + "sha256": "f9652ad6fad84e3d5d9fb2181e91e3e38e2a1ff3d6eec3dd18e76db489ce417e", + "size": 5790 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/agent/bootstrap.json": { + "sha256": "8a3916db45911667d182143670fe4edfb134d999147e568a8af01013ff22ca93", + "size": 9047 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/config.json": { + "sha256": "8363582d3a64157461c9d5ba41ea7f66ae8d353480bce033ec38d500849ded13", + "size": 1562 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/lock.json": { + "sha256": "8a63eef2c917352f843b69ec49cac0e3a6da52b6b0e8cf6b6a0daefc1a102419", + "size": 2091 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/result.json": { + "sha256": "0878bb05a625a7f36d0f90fd39f02e30e522daa6af9781399da7f34930b0aedb", + "size": 4464 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/trial.log": { + "sha256": "da82351fd19b477c88954744e38d99fe848cd8f9200d31a82096edf728670da4", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/result.json": { + "sha256": "bfb66a0e5008f1ce1399f00b11866bb79560c7094befb2abc15e15c22fbbdbdc", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/config.json": { + "sha256": "15418b0b04f857865688fe4c44b66477cf65f664fd480bfca8475a5be1c0ee22", + "size": 3403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/harbor.log": { + "sha256": "5c5a9c2979d65418e82ffffe2c9fa64d7c717d88ca3c5c5e8c8d6c6bd7c66fee", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/manifest.json": { + "sha256": "86d588aafeabba43bd931a297bc78b960761b2f9b878bece05301deab3defad3", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "7e62bf14d086f141e8de80852a2db1da4d1d1a9d8deb3c6c5abbac71a42574d5", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "94a9937fc2cfa054e3b5713419c968b1dd2e50fb91fb4180c1870efc7fb3eea2", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "de5e118341ab42914a4593a23ac283362e37171ff5d4e927736349a4bcba48fe", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "5815622800cb794d19b46bebff9376de28690fdfb616d106425bb9a635beffad", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "879287073cbba96602e9f936e1464d60e799856b36d0cfe033d84022d59943bd", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "be31d9d0b2e239cb4de0fc3cd6f448d8e8d99ebedcc086292030d69c12132abf", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/summary.json": { + "sha256": "2cef49e9735b839280086ea02251978e776fbebbde556d0bb5ecc2dcbebbaa64", + "size": 35014 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/agent/bootstrap.json": { + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d", + "size": 7379 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/config.json": { + "sha256": "bda0383c089c346df93f254ee98e466d44469492c04ba6606eea3ba679febcbf", + "size": 1566 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/lock.json": { + "sha256": "229a1c139c63aa78de941e548e01728f17bd391288f06f24c8c4a85b01c1c904", + "size": 2093 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/result.json": { + "sha256": "56e49dc1210e839e68a550346dd1011637a9780345c7a5a9cdf84fea3c360e75", + "size": 4476 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/trial.log": { + "sha256": "f0b8db055516b5b9385fbc86d04fc3687b5d7ace187ea4a94bed0f9acf62e948", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/config.json": { + "sha256": "7c4840cf5e2ca3d2495667240f4b7c19e50c83b746b07b823287a1545301dddf", + "size": 1781 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/job.log": { + "sha256": "aac0533b2bbc60cb44a01af9cd9073a3d245f520472a479a8e8b01ca16ed2557", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/lock.json": { + "sha256": "31de9000f21eaec55790ce8bb3c55d45505e20415e6902c226736872e5fb30d4", + "size": 5788 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/result.json": { + "sha256": "890c7ee21deec37a271f5f84b48f4509b202726d04b9aae996c4ebf9c074e3de", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/agent/bootstrap.json": { + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a", + "size": 21595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/config.json": { + "sha256": "dabae60c2258b325a14da0a4dbe2b3ced6783f2c7119a6fd0c7874e7675a299f", + "size": 1594 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/lock.json": { + "sha256": "40c8d1e9e4ce65a5da7d6254eff92cecda0b637d7385abf1212e6e9af3740a8f", + "size": 2107 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/result.json": { + "sha256": "a13296c3bbd5cfb2e2e6e7c0e37d360492178221e49a2cd868078abe876d7629", + "size": 4560 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/trial.log": { + "sha256": "cb73e5dd75d23028c6ef59b9fb56396bbedfd260806fc247057b564b050ae038", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/config.json": { + "sha256": "d1933ae0336e10772448feb7bedac1e22534c56399890a9bb97225339fd55e40", + "size": 3401 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/harbor.log": { + "sha256": "29143932e0a70485258462b8f7322068e0a0855b9af50c3a8d319c6d77b82488", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/manifest.json": { + "sha256": "2c81a93d2ba080cb4bcc3c359eb10e95b4074eb5c324f2076fb5db330fab8942", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "efd125fa2496428881579e58abfe0771e99e77d54d2a822657a5ffe61a7a61f8", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "40c2bf2d10ac8ce034364e100337a5b1b7e8219acf5da7c832ae30cf4c5cd7c2", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "30fd844405e109b5692e183e6ff7f801db8906ddacb725cc3f182f5df0d5833a", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "48dfaa5d270df3f8d5a72dece9d0dc61f6f8f2a09fb617b308d0401c978811d1", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "03e9ffe345c91522a949c72f19b06237f9d06e0fa6b79251560a6a5dd3f4185e", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "7ba13833538dc06c10665d44648c5a0acdebaf777208c384cbdf2a9d9b182000", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/summary.json": { + "sha256": "be4267f0ea72f63625f79ab7f8c43a598fb2eb358464564bda289a9b8e3db02b", + "size": 33132 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/agent/bootstrap.json": { + "sha256": "933708cbc5cd3c97f336c849d1eee8af9f3b824eddd6aa7506dc53f1657d8247", + "size": 15857 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/config.json": { + "sha256": "9d7fb837bbe44007b37c221cf9c4b100d22596e22e4ddfee59bc12aa46cad452", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/lock.json": { + "sha256": "de037361be733c1d43c23f7d15ba5a721dd4974ef411300ee5326954b9a2c108", + "size": 2094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/result.json": { + "sha256": "2a6b3a5ec5956d1249a5663543ac1149cb4a1fecfaadd0b7c34a6a1721939c60", + "size": 4482 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/trial.log": { + "sha256": "a7530dfcf982ac9052c4139e01f4ae046c6e9992151c4af93d3961fb1b86e238", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/config.json": { + "sha256": "6d45f91d19f21873c55016acbfe5c94ada67b6335928f085e13fe507ef788c98", + "size": 1770 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/job.log": { + "sha256": "c2ad73f6b9aa5ef9cb81969ec1e914b34343a44084ad057eb3d96425e5a037e4", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/lock.json": { + "sha256": "c0356fc4844fc97823ea046f02a935ed4ecaadadd2cfba1035cce7c96c90f509", + "size": 5777 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/agent/bootstrap.json": { + "sha256": "8a30f5515bd2ccd7749dc2993f8a83b0b668c109ec78febbf08cad59e3b4caa1", + "size": 2536 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/config.json": { + "sha256": "020e1890de768d5e3b56b7e70e106055f586f86f05c541bd8499226e36045ad3", + "size": 1570 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/lock.json": { + "sha256": "876bb3a20930940af5353b750b2f814bbfda9dac437a21004dadc7f5d6bb4e9f", + "size": 2095 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/result.json": { + "sha256": "06d54613223f16144e4d17d9c12e0a9061487bc4fffba4d0d146ef0e55cefdfb", + "size": 4488 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/trial.log": { + "sha256": "50b557ff3efef7ecf0aacb4a89958a9057880c5ccbcf93c04ff511933d71f4e3", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/result.json": { + "sha256": "eb5737fcc9c836197b454b1ef117c6f17c5b69b9c54c654ed0a29f547e4bf05d", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/config.json": { + "sha256": "582471c88f7186d8c092dbf56603f73faa6436ddab5ec82529ce77cf1488bdbb", + "size": 3390 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/harbor.log": { + "sha256": "84fe4a7ac4fdcfd794a91e006b9f28218ab5583778328de5a0c39a603e3d268b", + "size": 1048 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/manifest.json": { + "sha256": "757978d6b034758a75027aac61264ae6eb72243da9338ffafdeae28cfa78051b", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "f11b36b4a034a2d1d43001668e975f334d85e7f6510703202c5f8a7fc6a11054", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "f1499f82c01deffa75fe7cd61c88e196113cfc6e4f7ba7bb0896a22204ca59a2", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "a984961f6d8085b387204c17e7f6519829ca0c473e07a40b8aeae33c74b5e4cc", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "8931aec1611c66d55ff0e7762f3cc496b8e404adfc1b47575e0cf14b83a3a04a", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "633a2843dee11dedad275ddc6c9c98ddab41d6a5740604b95b26f8b7d205527e", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "9e69ca191307c2336b3f63b0aeee483128bef63e6ff67e8046633b554bb950c1", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/summary.json": { + "sha256": "40055254b4c8127df523ee89e946b0426f1f65b2935643c17bb447d5f40f3a5c", + "size": 21275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/agent/bootstrap.json": { + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d", + "size": 7379 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/config.json": { + "sha256": "503b6906706eec3c8f03e997da474bb8cc461301ebbb6144ca79b4617bd831dc", + "size": 1564 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/lock.json": { + "sha256": "204ae66ef37f5ce197dc4ceeaa786aa8a50d478cd10fe7830ad9fa73a7a11757", + "size": 2092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/result.json": { + "sha256": "f9a470171cc88e6a3728a04b99d9ef0d9abf92d9768cc9e63cbdd93686b1582c", + "size": 4470 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/trial.log": { + "sha256": "ffa4e0cf68bfdaa003716259dd64bc8ce1e78bd9a174f3637bc52b33a87e4114", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/config.json": { + "sha256": "addca9b24f7e9378d7ac8b816f55ba269e0cc38cd72915a77c715bc391055910", + "size": 1763 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/job.log": { + "sha256": "5a5b47fd20cc1279634e8607bbb65fd2b0b4fab4b240c4ad50a24873ffebbefe", + "size": 13646 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/lock.json": { + "sha256": "29f91e15644f0dfa884271dd1d2a7b9e3e048246054d6228777cf2b3d058fdb5", + "size": 5770 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/config.json": { + "sha256": "ebe0d31ecd4de147448feefc054c5f4d5e1d783ed01e31ac67e5dc6ba61c8061", + "size": 1560 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/lock.json": { + "sha256": "cacf7efa401a49154bf5e7d07ed4b24158cd24cfda5006b2e5e1f85762a8cc7d", + "size": 2090 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/result.json": { + "sha256": "a0d6f78c049b56f4ca2e954e982800a488134b1c3b3af8b1fbe10abc701e386d", + "size": 4458 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/trial.log": { + "sha256": "290462078f362d3aa0bab2bcf5869e9a96bda880459489795a77c7af256e924b", + "size": 6823 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/result.json": { + "sha256": "c7cfd3c38840148f507826d9879976e8e4875cbdd19db26f26af083fccccf4cf", + "size": 947 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/config.json": { + "sha256": "480dfeba1d5b7d2f4ff91fe78df52bb95288207e10e2f00152916f1a57205be6", + "size": 3383 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/harbor.log": { + "sha256": "9632ecbe01d9fab1a94abb01053a0096ea7982ae7b43fb204532626040256e9f", + "size": 1046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/manifest.json": { + "sha256": "f8e99d12bdfa6699177b64203869a4cee08d713a5349a251f445a60eb6eddd80", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "60e6ae2bc1d285998a01993f37355b1341cb84008c859b18206ce2a55ad9d1c6", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "f971ccb9bf73bd1c4561c101321f5005cabba84160f7b5d5ad66629aea8d821f", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "170bbd389fdae4ff21b6db6199cf05f26bbd26b644a0279c6e2ccd4c161e251f", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "a8f57d1218d589eff90b8c2e77d2503d9806b1f0e9c176353148ab4f6400b4fb", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "6c72d531ac65c8b11d4ea59f6a848d2a925fe469ad2083f2ad01c839e8f23cc7", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "3925e5c3d4e9f24cf4165e72907902e683d1d9fc5e1dffda9f4562f438527e93", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/summary.json": { + "sha256": "32a7d803060d971c6db5bfe036524455d3000efa4997f37f4e8887acc1d62e32", + "size": 33225 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/config.json": { + "sha256": "aad3806b07759f9b7671bf3779d5ed5d68c740efc53ed5216441b2d4b95fb6e7", + "size": 1546 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/job.log": { + "sha256": "1a210281a4411e8c89d7e1abf9f70ed447dab775e13d3b58bd58bb7cbe50ebd8", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/.anonymous-user-id": { + "sha256": "f7050308448b2c27f36583299a357c3540696b8f20c4d2b5eb4a23813cc8accf", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "de54531c0b2f287c81c7bd4bb187a7d084329181caa6f32cc68017f714687fde", + "size": 52244 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh-events.jsonl": { + "sha256": "24ac5c6765b18ad9b0598642b3a13ac7edb05f40ce7fc9bc18ba965a2a08de49", + "size": 60625 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh-result.json": { + "sha256": "c1bda054daaecefe6c4a8e812194245a7c60353b9099b8476682f4f5e8ca45ae", + "size": 567 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/config.json": { + "sha256": "615ad66190c8a5318ca5159b84172a675d8210a9bdae6906d7d2e91810d46dcc", + "size": 1511 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/lock.json": { + "sha256": "cf3dcf80dabf2c0b574ebae2f4bd5323aa1e3b070e308392e5c9cb5ed37d21d5", + "size": 2094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/result.json": { + "sha256": "aa9e6b5d4ac06b90685252efbb142dc8047f6c9f7b764138c39e6028541e0e35", + "size": 36419 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/trial.log": { + "sha256": "1a210281a4411e8c89d7e1abf9f70ed447dab775e13d3b58bd58bb7cbe50ebd8", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/attempts/1/ctrf.json": { + "sha256": "787209c91cb9ad2f0133a78244a0f0e23023e2bd701201a15ef549bf27aae9f6", + "size": 6555 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/attempts/1/test-stdout.txt": { + "sha256": "7fa5fc2c891910c002d4cf8836dd23b2b8e44fcae77018c6c4f9c9d9a4163de4", + "size": 8135 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/ctrf.json": { + "sha256": "787209c91cb9ad2f0133a78244a0f0e23023e2bd701201a15ef549bf27aae9f6", + "size": 6555 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/retry-summary.json": { + "sha256": "7b1e40b2c71890f9e38e3595c850e13371db3253622d3c6112b47dbdc3796108", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/test-stdout.txt": { + "sha256": "7fa5fc2c891910c002d4cf8836dd23b2b8e44fcae77018c6c4f9c9d9a4163de4", + "size": 8135 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/lock.json": { + "sha256": "e3704ce625922ff623ab7452b2ec61098f1bf555adabc19aca4cad1b8add6a42", + "size": 3248 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/result.json": { + "sha256": "47b7938e53709a4f5d04e532ecddfc6daa48b9ac9947908d554bc400ad999ced", + "size": 1132 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/config.json": { + "sha256": "e0b670a59da7c8751fa10f2489c0730b4a1c89f1d49479f14de51d76ae4391b3", + "size": 3090 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/harbor.log": { + "sha256": "fd00ee532b14b7135fa21ce5a16130981e3718d94575dc947c140381a705b16b", + "size": 1261 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/manifest.json": { + "sha256": "2b8f9822250021cd497f3536b485da8c104e6cca9d80a49ddde7820745f8deaa", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "064efef6452c7ebb22b9a4d8ee4e9f47da99bd7eef1716440ff4256ab2983731", + "size": 582 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "8afc5445b2deae574a5e7178c7444969ded4db6e8268f3b5d3b8484f7a34ceb9", + "size": 18701 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "e27c0b2d3c4017d1e08c298a0261ecf6c8d6b687eecbe288006b3cfaa3680d08", + "size": 12009 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "f6e533b11751823e911eca6d2f834dc080377b874dd2d17ae45c202dc7e1e172", + "size": 15043 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "9f1eeb36dbe71941b2a354b2c79e0877aaa2a719a024c8420ec83e1b62e54692", + "size": 18392 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "f62a083a2a97ec0167f463fbacdfd0b4f97e770f9e06db38fff423e109117512", + "size": 14691 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/summary.json": { + "sha256": "bf55c882abad4f9388f6bb11f7c2ad4b2ff0fec9b122295ae926c0440aea6b88", + "size": 25035 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/config.json": { + "sha256": "24eac07d39f66318bba3aaa9b381668d428fb4a0ff46127abd3b517bdeaf0839", + "size": 1584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/job.log": { + "sha256": "631895d16f4b14c398863ef2b5e1f81dfb5a83dbd25e07996d3063e4d090816e", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/.anonymous-user-id": { + "sha256": "fc399c86d22a224e13dee759b791b4387738f95679e901735e9e4b2ae81de4ae", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "faf0aa25924f09e4b62f094cfecb1cb78551e69798f9b006cc6fb80a048c2b98", + "size": 743686 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh-events.jsonl": { + "sha256": "cfe9604b53d98704bdf0088e29c85925a4c3bfaf545a0e229ff7eecc6b0d2586", + "size": 834651 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh-result.json": { + "sha256": "9665a3b8d679e8fa420d777941e344668f0d006393a6ae6dff32a7e26ffe216d", + "size": 746 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/config.json": { + "sha256": "13014454e3ff71affc417116088e206e76d89b5bc5cda50b6eb6a6c9c77351d1", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/lock.json": { + "sha256": "e5a2993be5cabada020d76a4e637b344da90fd5361b002374e9e14fe9fef9959", + "size": 2113 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/result.json": { + "sha256": "6e1542ae91cf1eba19ff628595fdb4922bb487ab10dec811c9af61f8c8e0a10e", + "size": 36576 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/trial.log": { + "sha256": "631895d16f4b14c398863ef2b5e1f81dfb5a83dbd25e07996d3063e4d090816e", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/attempts/1/ctrf.json": { + "sha256": "71a1350b2936613ef59837398e0a4c171a49f20794cd162ec12fef6008d191f7", + "size": 2547 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/attempts/1/test-stdout.txt": { + "sha256": "c1b8385af6f24121316013915f21817b1b0bd8dcee6b8623ce60e217ea5e880f", + "size": 1504 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/ctrf.json": { + "sha256": "71a1350b2936613ef59837398e0a4c171a49f20794cd162ec12fef6008d191f7", + "size": 2547 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/retry-summary.json": { + "sha256": "4e304dd6218d48212f724c93b1b2c5fc1aff4aca1f64ef748aab6d8782397aa3", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/test-stdout.txt": { + "sha256": "c1b8385af6f24121316013915f21817b1b0bd8dcee6b8623ce60e217ea5e880f", + "size": 1504 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/lock.json": { + "sha256": "65cbb03d6ab348fdf4e4a8a885f7698eb6e93267ad8950cb9c04c69efb743856", + "size": 3267 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/result.json": { + "sha256": "9b31072c941da7712e9ce6049ed394ee3201eff145569d1051220bb73994ae0b", + "size": 1156 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/config.json": { + "sha256": "b0b250afcac986756f36b26dd7e4c82af7cd8d5d827ce1e2fada7b573e280aef", + "size": 3128 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/harbor.log": { + "sha256": "c496d3a388caf7f0a35ad602e3df571162a3301e6f4fbbac95f23fd8c04037c4", + "size": 1299 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/manifest.json": { + "sha256": "d1bfa964cc85d0f4112bc2403a619f4beee2da7f861c9c1fcbcea8c2a2c36086", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "f71343019914415f205b6e250155cd8ccabbe5a23350f098d2308b953e108318", + "size": 601 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "9abff74a2bd3aa1187eea835e73413286ce9215b13dd4c85db78e56cc138b100", + "size": 18720 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "7a03d69ff76a42aef425b78026cd0106f6ed176fbee289036df7a142ff09b0be", + "size": 12028 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "4ab101aa580dd6cdc20a6344e28c3f64e3d2918f62542b538ed2642bef3373d9", + "size": 15062 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "e54ae20191a700cb018e00d04286dcf4db54dd683f8a1ae1b361c53507ddca45", + "size": 18411 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "d5194f5dfb153f5197c30fb645d8e2b9979cc7d1ef1d35f088bb4f5e72302df5", + "size": 14710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/summary.json": { + "sha256": "58ac0a0fdcea0dd4059e76eafb873fd26469141a0ba3c8c68d09a0d213d54fc5", + "size": 25074 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/config.json": { + "sha256": "3de39475758019523a03a055d3a13d8dd12ff1b3afb3a1ac32fb652ff5a5de0d", + "size": 1566 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/job.log": { + "sha256": "f861de47c1341271544e2728e1d834e8967ed948c6cf9276acc774b4cfcc822b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/lock.json": { + "sha256": "4e544ce4ca183c73b388ab354371c429a9857bb21ab256b032d0110dbec8746e", + "size": 3258 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/.anonymous-user-id": { + "sha256": "922c74811fd8fb730fe2f8a8d8682c2accbb2584c950e636be3502e56cda1d72", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "a8b27dbcf4c9e2c8243c67e38a75158b75c1f1f8bd11092a3a3d760bde8d0ad3", + "size": 64257 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh-events.jsonl": { + "sha256": "4030669235c95d1af8ae4ffb456e23480867f9e30e81fe480e58a89b13100377", + "size": 71138 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh-result.json": { + "sha256": "31073c4d550483ddd425dda3b1b05e368eb02daeda1ec9b845ec28049a0f849f", + "size": 568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/config.json": { + "sha256": "b3e80b267378fc0902fc3ddf661367cb2717f95c6a63ca5ac57cce74c58a78d6", + "size": 1541 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/lock.json": { + "sha256": "ed35fa367818ad1d42d46ca4fe07b8a7dbf8766ea20e9c6ebd9a9c30eb252727", + "size": 2104 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/result.json": { + "sha256": "d02b44a1b25ef6fa02f8986f5bcebee4d587b77b6a74f8dfb1b95dba292211fb", + "size": 36499 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/trial.log": { + "sha256": "f861de47c1341271544e2728e1d834e8967ed948c6cf9276acc774b4cfcc822b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/attempts/1/ctrf.json": { + "sha256": "09e4f8cb35cdc4388ff24a26496a7cdb3f3fd3de204a5239a067ae4db02e96e3", + "size": 2451 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/attempts/1/test-stdout.txt": { + "sha256": "d9446e5503fc0cea5ca1b2b061c0bc84c9a378a1d4c34ab8fc59666285b575e7", + "size": 2586 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/ctrf.json": { + "sha256": "09e4f8cb35cdc4388ff24a26496a7cdb3f3fd3de204a5239a067ae4db02e96e3", + "size": 2451 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/retry-summary.json": { + "sha256": "ffd57a60e6e19e4bf54d33c853f9010a43030bef21612dc1d44fe81433c43e77", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/test-stdout.txt": { + "sha256": "d9446e5503fc0cea5ca1b2b061c0bc84c9a378a1d4c34ab8fc59666285b575e7", + "size": 2586 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/result.json": { + "sha256": "7be1cb78c8bed938e862a646d826a56ad369ae03a31186bf370361559c8e4fe4", + "size": 1142 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/config.json": { + "sha256": "94178d924b996009ef571fa6bc69301551be952b3bae9ac66175208ec0a41aa7", + "size": 3110 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/harbor.log": { + "sha256": "430332e17aedb663f05c3bab64de458f796e31418447a748fef6da413b932ee5", + "size": 1281 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/manifest.json": { + "sha256": "fcb2bfde554aea943ae7ea256b857fde7e0f18a29d452fa95289a246e474fd52", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "b10e340040ad7cbf3015d827b3df11337f72a43789f64c97fde5558e5be8a3b5", + "size": 592 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "ca814499b42c38186c1bd26e5b8f454c4ca03ace75b3c9f9be1629ee3534c7d1", + "size": 18711 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "1f63df228dfa8b85480e6a65898848e99033af8f03ff60e674f7e06f532b1383", + "size": 12019 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "2840f4aefa743feb4a03207b038920c0c8ccdb636ec1fad6505474dfbe99aba1", + "size": 15053 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "b894999a7824b9006704ba4ec3ef71779b8cc6128893451b01868752874d1384", + "size": 18402 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "98b5ee6c0b9b53c52373d09e831ccb9862d8fa5e01d1e47bbbdc36f684ca647d", + "size": 14701 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/summary.json": { + "sha256": "2afb4eb373ef5cb4765f237da4864e240185e14ec16b2e9263d6fda6c4332aa0", + "size": 25056 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/config.json": { + "sha256": "f6c3a00bd17089dd370ef8be1d0c578f70e89c52a4dbcc697c222ccff8191ce0", + "size": 1566 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/job.log": { + "sha256": "482d8ef29904f34d520f3d1d778839550a22fe5bb7b91f9b4367c7adac3c198b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/lock.json": { + "sha256": "f8f799dbff1ee116f1451f2d33a2b505c9aafec0b1d5e5b585c36e3daab15e66", + "size": 3258 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/.anonymous-user-id": { + "sha256": "48bf5363a861bb9e17cdc976e01da3c6aeab9c859c50c520056e883c818ad989", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "45bca7a942e77ed44a678813754a0149ac59c96959e03eb09b2ce8e13c40a943", + "size": 270571 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh-events.jsonl": { + "sha256": "746fb7b6120b33b3ce2d6225c56fa9738a634586bc441f0abd9f08b949a55490", + "size": 304694 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh-result.json": { + "sha256": "de424222d6425b58b317547251f729d19c563a621f7fc324db4aa3b124b27d41", + "size": 622 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/config.json": { + "sha256": "199704138ee162c6d843837f8d4a9b0271ef8bcc035f1b55c11d5d4caa1dab39", + "size": 1541 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/lock.json": { + "sha256": "5dc6aac2abe88f886ee7acecb38471d86c40055fe7ac541e721a22818dd74810", + "size": 2104 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/result.json": { + "sha256": "a863044a3edb62310b85043a6c9cbe76f907427566b4fa463e6b77dc8bd7ab11", + "size": 38497 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/trial.log": { + "sha256": "482d8ef29904f34d520f3d1d778839550a22fe5bb7b91f9b4367c7adac3c198b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/attempts/1/ctrf.json": { + "sha256": "f099ff03f48ca015726438a9fe7fd73f54645d3f22fc58a347a3bb8071d6057d", + "size": 2169 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/attempts/1/test-stdout.txt": { + "sha256": "71ed332ca5614ceb6b4267447b80d71a048b7d55a5f50e5a47b987d84ba8732f", + "size": 1573 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/ctrf.json": { + "sha256": "f099ff03f48ca015726438a9fe7fd73f54645d3f22fc58a347a3bb8071d6057d", + "size": 2169 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/retry-summary.json": { + "sha256": "e41419110d4702693b481e9f76ce2eede341dbdf7c9deb84fd3745eaa03b3163", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/test-stdout.txt": { + "sha256": "71ed332ca5614ceb6b4267447b80d71a048b7d55a5f50e5a47b987d84ba8732f", + "size": 1573 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/result.json": { + "sha256": "13a374397ec4e3febfdd21ac0a9d72394fb773f91e2ad21630433ae397c569c4", + "size": 1144 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/config.json": { + "sha256": "1ae54d8b5c7de6aa5e58cf9974b4a3747b895de05951dcee7f21134b0685063f", + "size": 3110 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/harbor.log": { + "sha256": "8bb9577af63fb30091d4905194aebf14064bd61459aaace4f340d1be8361f2f7", + "size": 1281 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/manifest.json": { + "sha256": "fcb2bfde554aea943ae7ea256b857fde7e0f18a29d452fa95289a246e474fd52", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "73a24c41aec9f2980f848a5583aaf60776f42ae030b3f1273ddbf647282f7254", + "size": 592 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "d258f05077fdc4d408b016349def079dfb4c5311e74c8c2bd3ef39dee167d6ee", + "size": 18711 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "ca401837f9337b5afc93e8e81caef708b8233c7dbe9c13dcef9916d36019ad00", + "size": 12019 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "53d6616719432d136c95312378a31122fd7464a29a575628cdd13887dfa7bc6b", + "size": 15053 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "0453fa8dbdabda40a972b3b315f0520b89b4e461466d823eb2cd2d29e7231481", + "size": 18402 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "a9aebaa7cb81eb61414c9abb35c7e08e9e9091b07101f19fb1a802b0f64424a4", + "size": 14701 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/summary.json": { + "sha256": "8ae84f57c1cbb0bf1a1732831dedf5bca98bbfa101c96da1ac007bc8738367a9", + "size": 26732 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/config.json": { + "sha256": "7fcdf1db4448557157c3cbb30a4bf900b2be64dc5985e3f07971b5efbf255041", + "size": 1576 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/job.log": { + "sha256": "2b084057a8feb1d57b4c4f71b410d13b5608d4fe3a017a6a04bf6945f4e821b9", + "size": 12897 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/lock.json": { + "sha256": "eb7994276a6a3f9933d9f456605cbdabb672f4f5385971e5da8d7abf07326957", + "size": 3263 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/.anonymous-user-id": { + "sha256": "01c81a86404b8eb28f8d9e9c3eaa49dd8be0513285685594b89da4922e5eb52f", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "ba074f750d4ad405d2f596c9ef2e2a09fdaf6d1adeb34986ab38b39d86974296", + "size": 898480 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh-events.jsonl": { + "sha256": "a4bfa78a56640ebc50e51ef50d16f6ff84be22f9edca084554d85e3a6503ffa0", + "size": 1011218 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/process-supervision.json": { + "sha256": "ecf9ddfe1194866091d36f7817a57689da8cdd81ee2df92be1c550c2f3edddd3", + "size": 75 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/config.json": { + "sha256": "3c406cfbe0bb7b9759e66714abae88cac45e36cfce4fa9c8dd0b084b21d68209", + "size": 1556 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/exception.txt": { + "sha256": "1fddc8f4fe3c49c56a7d2c5317f701ad6e340b3ad36e910d18de5dda5bd01bb1", + "size": 5061 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/lock.json": { + "sha256": "4c3ebb4dedbefe8c8b654a6aa3f8451211739d62d1a6bd7929a95c44dfbd5bee", + "size": 2109 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/result.json": { + "sha256": "e934b2adf693122b884894db9146c87ba83224b684a05c3d5a081d9f40c2e855", + "size": 41994 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/trial.log": { + "sha256": "4ea89b62476946bbfde37ae672aa15f401b6a1d9df45c40ad01a7e5907b529bd", + "size": 12710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/attempts/1/ctrf.json": { + "sha256": "b93ca5d9a87d1bd312793c87bf89d6f0cbc896a85e49659d9522d965e83d579a", + "size": 1421 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/attempts/1/test-stdout.txt": { + "sha256": "5b663727ff352b125abe6e2ce9219cd9e086c78dd203754b7aed5f27e132affb", + "size": 2014 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/ctrf.json": { + "sha256": "b93ca5d9a87d1bd312793c87bf89d6f0cbc896a85e49659d9522d965e83d579a", + "size": 1421 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/retry-summary.json": { + "sha256": "05d636ce9058e41247dd69cacbc748590b45eaea1cd509e40e88ede6c6048a83", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/test-stdout.txt": { + "sha256": "5b663727ff352b125abe6e2ce9219cd9e086c78dd203754b7aed5f27e132affb", + "size": 2014 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/result.json": { + "sha256": "e82c1c1a2e2eb6a3501006a1da029a685c3a73a2790cd6e88347850312406757", + "size": 1296 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/config.json": { + "sha256": "489009bd9f3164fc507508ffdc3c2e5ed87b48ac766019716cdeb4e444f81711", + "size": 3120 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/harbor.log": { + "sha256": "609a6b437a7064fe860e36a4f9f293476786085bf7e45d721b95cc3f5e334c84", + "size": 1629 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/manifest.json": { + "sha256": "a9394befa66cbac76b4ae051b0f95c936e12d7c7aa628e1b094927bd47895358", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "7c396e8f1ce368389dc56295712d99f1905eda00ef266c368382f8fcf23a2658", + "size": 597 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "3535c8c13471ddf3b796c7f478613e5e405819e7502c1d1b8fd46f26a542bc18", + "size": 18716 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "f3cd751d4f302c56a07f770310138c3696190aaaf9a5e486a820d5e71928c76d", + "size": 12024 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "ea3490ff849a48538da36435c75a97761b81864e5e2d3acbfe7137b0913805ae", + "size": 15058 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "ef80336383efc64a81850aeac3d0db65ebede7c9190e8b790cc2bfe5e0c20068", + "size": 18407 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "ca78577536e431409891f2ce4f5e92c62b13e20bf8b36d89bb40ab9791e3a368", + "size": 14706 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/summary.json": { + "sha256": "5caf1128423df6eb921556ff38df66ec81f6995127a4bfc7e2184eb136da9a77", + "size": 25063 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/config.json": { + "sha256": "21998293b116a767f47d5314ccb666473226f9edd1192d761f87d9e736043af1", + "size": 1552 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/job.log": { + "sha256": "e99a13c9480b90639ef4428786e692db240fc0301cd1d96745df8a31ac296b9b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/lock.json": { + "sha256": "37a5e0b0d825a08f6a1cbeabbde9f33b5ba9ba84fd0765da4c67337fbcf6c2d7", + "size": 3251 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/bootstrap.json": { + "sha256": "8a30f5515bd2ccd7749dc2993f8a83b0b668c109ec78febbf08cad59e3b4caa1", + "size": 2536 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/.anonymous-user-id": { + "sha256": "7fb3fa102cd7133e6f6b83298057c8c7c4c6d4ed065666ef80e045a0fe8f9929", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "0dbe8542200ecc8e6305494b3509ec245e240202621e1666a2a421b1628e2a73", + "size": 929129 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh-events.jsonl": { + "sha256": "de54ed6deb6e7029b2f805b488e76b83df79ac4d543a423dfb6840e665c42285", + "size": 1036204 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh-result.json": { + "sha256": "de1e931762ee6a3f6c58364ec36ce5cb295650f76cf79361a2faf930cdf0bb81", + "size": 126 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/config.json": { + "sha256": "3fc165b37c73d52715e826740a898bea6dc484dfb04828503e70e2a6b4dcad12", + "size": 1520 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/lock.json": { + "sha256": "a7b99859e77af2b37f5af24447f9f390131a6eef1f550e9054fb88d2c65abd57", + "size": 2097 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/result.json": { + "sha256": "37676ebee576524bf8103b0bdc29c97c587f954d58122958ebe0983eb3b686a8", + "size": 9269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/trial.log": { + "sha256": "e99a13c9480b90639ef4428786e692db240fc0301cd1d96745df8a31ac296b9b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/attempts/1/ctrf.json": { + "sha256": "434f04d5502e60cfe742cf747fcb2a71412437f453ed76760d167593e9628b7c", + "size": 1093 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/attempts/1/test-stdout.txt": { + "sha256": "f128f0efc5ee5a556e06100cacd777393fbcc796c16bee8df99a01271715058a", + "size": 1260 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/ctrf.json": { + "sha256": "434f04d5502e60cfe742cf747fcb2a71412437f453ed76760d167593e9628b7c", + "size": 1093 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/retry-summary.json": { + "sha256": "048f14f0da7b01d7574021ca3d4613441ad00683c40a2af09ea0c79225329356", + "size": 271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/test-stdout.txt": { + "sha256": "f128f0efc5ee5a556e06100cacd777393fbcc796c16bee8df99a01271715058a", + "size": 1260 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/result.json": { + "sha256": "c4761e5c2c9ad1bee7ab74ef2654ad336b53bc83f654389cbbe3386222a8ef0d", + "size": 1140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/config.json": { + "sha256": "d9ec1602d13d24b36373d4f9655f136a4fd316e1ea3670b80a9ed881ffea4804", + "size": 3096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/harbor.log": { + "sha256": "a3ef10de30d0458fb2a4b1f049388d55b65725ef048c8e377fed79fc03b5637a", + "size": 1268 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/manifest.json": { + "sha256": "144d39b5e20bf5333fa90dbcb759eb8a8b78833460b856bde474236828cb8bc3", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "f2a380bc1127139791240ec64e54cf7de68853fec98ecfe0970c9a0d3b15690e", + "size": 585 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "f5edea8ef2415d446747b919cfe3faf9d5e732f9fd4802f5a38bf53ca2cb0c7d", + "size": 18704 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "c3262db887758d84e59010b06f1210d2f77f9fb41ff6badab50c75a2cf67a8d1", + "size": 12012 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "a215487f9d425323417eab555f43d279c566092577a173c166c9c7e741696d20", + "size": 15046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "9ec90c558f1ba9fe9d4895aafc816f232c8d9ca496b0f23d379ea289b0b02bab", + "size": 18395 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "74deccdf3fae3d3eb1d47b609d9cf9c5fb5bcd277ac84cfd10b29503c13140a2", + "size": 14694 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/summary.json": { + "sha256": "ae896c1b57e9b640272382044a3cb30b6b2815b9820a2bb0f745b00d94cca0de", + "size": 3587 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/config.json": { + "sha256": "a26e2f005b4f2f4a331d2247a29f5fe9e78257c8a9a31396cfd79ed45c1b022d", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/job.log": { + "sha256": "0b70cbdaf5a0e27bd10c89eb37657b1453cf2c28cb6d36efba6de1746783674d", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/lock.json": { + "sha256": "bf94854a50c026fc61b2add91f444c0794a514097aeba407de2bd09c97ca9040", + "size": 3259 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/bootstrap.json": { + "sha256": "38a018af813827b21bff5685e056300a44dc7899ae803e02ae564568808be3fd", + "size": 19659 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/.anonymous-user-id": { + "sha256": "64db700bfd82dc32955968e93068a5f2e306c1e410e2404b5dd6ae582362a9ca", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "a925bd0b4ba595a90d60d9cdd92e3efa8128e6dac564bdf62de68fa221de3f74", + "size": 89084 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh-events.jsonl": { + "sha256": "5952cc8a6ab085f0160b13e29fa868f47827d01bb175ef51b1e66c017fee3b3a", + "size": 98358 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh-result.json": { + "sha256": "e11b06159d9a0379f5d3997cbf4a2a6d8630bd441f8272eec1e08740ff2cb2eb", + "size": 650 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/config.json": { + "sha256": "6a66a2fcb26cc2fd03b3818370cef0ad1ce803aebf7c8dd3163619ad3867af33", + "size": 1544 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/lock.json": { + "sha256": "bf10b1909588763a36bcbfca9968252e905eba70200528e09891443dbf839c31", + "size": 2105 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/result.json": { + "sha256": "8d3dbf7742c155f83ff42673a35fbdbf336cc76e6014e7b44425f3aa36464a56", + "size": 33700 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/trial.log": { + "sha256": "0b70cbdaf5a0e27bd10c89eb37657b1453cf2c28cb6d36efba6de1746783674d", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/attempts/1/test-stdout.txt": { + "sha256": "790fab06d71b6afc57e26aff4a9ff7da1df45708a2dc0abb5b6c1a4809d7c3d5", + "size": 1002 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/retry-summary.json": { + "sha256": "aaa5241bdfee3617dc074356809311cab05e39bfc565bcfcf6c648b7e6ea377b", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/test-stdout.txt": { + "sha256": "790fab06d71b6afc57e26aff4a9ff7da1df45708a2dc0abb5b6c1a4809d7c3d5", + "size": 1002 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/result.json": { + "sha256": "22fff7c924a1e66c7cbd7734305d2c6150928d4d2e8254cc5beea943b0d87d32", + "size": 1141 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/config.json": { + "sha256": "8e8d48ac943493b44c102825fcea323bd6ec9b000a09d2370878278789f4b518", + "size": 3112 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/harbor.log": { + "sha256": "84a48589e5ef46c692de0e1bfc72008c50875da104bb7c1ddd4e6411d0a01891", + "size": 1283 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/manifest.json": { + "sha256": "6fa7bc807d925fbd687aca99feda437bfd33a0b1daf801b71266b0206e32692b", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "5da7c87487cb929f899dcd301d0b764f981a7f162ed8c6a96a5360c2c8434827", + "size": 593 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "05860decb28719e8430de60d44e2bf4611d66350308b574466c5459737c8416b", + "size": 18712 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "0ce4a76cc18176b209e9bdbb6cba46bfcb884c45b348439c36fd69a656f8908b", + "size": 12020 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "0bc5b49ffc18170b2377e6b2b2aa34078a35596cbabf9fccc4e8e757e241cef1", + "size": 15054 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "dc5e60de7ba72587c858ed51fe5df5a1cf1d855c432bba1f87f70b30bd13d500", + "size": 18403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "41781a65b5ef21f5bb616632eb8dec7c38f8bdfa182ace78f5176db5cb69dbaf", + "size": 14702 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/summary.json": { + "sha256": "4df7fee4104d91a1511cf74b96eb56584af8049c1f694251d620eacee3a13c82", + "size": 22783 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/config.json": { + "sha256": "a4b529de2523cb8b13c48e7ff9b559640ba31db138884f37afa52917139c46d1", + "size": 1562 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/job.log": { + "sha256": "d3e45b831c32f8af0b007413969b33165c2cd8ac12c668cfc7b7f9f7f3d06fc3", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/lock.json": { + "sha256": "2a642696e3a44a4e4845ab21b4465558977ada98efbeaceec0e3d4650d20f53c", + "size": 3256 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/bootstrap.json": { + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e", + "size": 15490 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/.anonymous-user-id": { + "sha256": "b123e714adc7a66ce0e0ecfb29da2984d3507161e17f94cbd1595c39817bc47d", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "23c57265311080c8c10570109f74cd82b0fcabc57f1bc4f50d11b4aadd9e0a3f", + "size": 77824 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh-events.jsonl": { + "sha256": "366acd57bcddd66e8415c0b9e6dc2d3bcb4cd40e755f9668de001a1cda0ab166", + "size": 88043 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh-result.json": { + "sha256": "508305c76d7fe4c41975e969fde913b418cf70e59d298c978a2069cedccf4d9f", + "size": 877 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/config.json": { + "sha256": "8aeaf6347086a3e1b6bacf54ce2baea38c0bcb08532b26667a5041d2c6ac29d9", + "size": 1535 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/lock.json": { + "sha256": "3eeaacda54d31c46d7d6e2783872bb46d190aff757d96489b52b1950a2558da7", + "size": 2102 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/result.json": { + "sha256": "d51ec031edb07f74bcaf32527fda2fe19a88526868709c194be3740fb1e84add", + "size": 27733 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/trial.log": { + "sha256": "d3e45b831c32f8af0b007413969b33165c2cd8ac12c668cfc7b7f9f7f3d06fc3", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/attempts/1/ctrf.json": { + "sha256": "fffec6cec8fa508ef105bc423ecdbddbca7099d213cc25a035fb184d66a37c58", + "size": 3185 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/attempts/1/test-stdout.txt": { + "sha256": "fc0078ba84a3973a5b78789011cad03edfd96c4485e85f54b4869f03865055d8", + "size": 2265 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/ctrf.json": { + "sha256": "fffec6cec8fa508ef105bc423ecdbddbca7099d213cc25a035fb184d66a37c58", + "size": 3185 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/retry-summary.json": { + "sha256": "82eb40e1ac90262fd6a9ac5f1401a3df82724c2f47522c2361867acba073a562", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/test-stdout.txt": { + "sha256": "fc0078ba84a3973a5b78789011cad03edfd96c4485e85f54b4869f03865055d8", + "size": 2265 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/result.json": { + "sha256": "9ba90a93091009356c45ec88033372ca646e4dc2243c9e64f76c91b202488be1", + "size": 1140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/config.json": { + "sha256": "f62567d789a0dce0e0ae922b869bcce1bdd64cbd5d78d1f9479fef47cc5ccf78", + "size": 3106 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/harbor.log": { + "sha256": "fa312e9adf1769646fb51714379142217ac889ff3277a5cbd4b7ca3a6669ca2f", + "size": 1277 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/manifest.json": { + "sha256": "39b824fc7941d63fcb1db8e9cd0f9bf7ffe6b852de5200e9d03240bbb409a20a", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "4472d9520521259ee97b926d43d0687d99ab777e9e6aaad147cea8d7d2b11406", + "size": 590 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "781234f3f3295a304a3f8046a42bd42cc2d3f2497eb1bad834210616b62ba63c", + "size": 18709 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "eef90edcf806c691eea9494bfa14901465e702b4745aaf6633fb90f8524c7670", + "size": 12017 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "3f0ca020beb2a4f6d7cc0c0096ff135a1f452770e56f490db1d9757b70659daa", + "size": 15051 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "c6f8fcfc316c204439c08e36822f5296b3665a59a95e1f57d82c1da06b9214d2", + "size": 18400 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "98efa6f3930acb0c9d081ff0a76408550c3cc7fca48746a1a940ce311b8fd021", + "size": 14699 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/summary.json": { + "sha256": "1450078b31e1d7f69f22a43c642be7a57e9aa6c21d11a7760928398d4d6de226", + "size": 18104 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/config.json": { + "sha256": "c099ef5ab183bb8380382345596876c2baead0cf6ceaba3ead26733d7a628047", + "size": 1566 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/job.log": { + "sha256": "bbe49ae9847b03bcc835b6fd70447910a731e117541c5ac67f49b50b86f6ac82", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/lock.json": { + "sha256": "8fe6a4231ebbc1cca925bd9be8c310c91c45e792619d06b9c5bab38004e51c4b", + "size": 3258 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/.anonymous-user-id": { + "sha256": "38489e247a3c2a356f06576223de8e61c0b4ca2d87e7aa12ee9f4e4e76b843e9", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "f9b54d1c61ec3af8af09da19fa02a66fb02c85052e5eadcc1306c0f3e3003aad", + "size": 82867 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh-events.jsonl": { + "sha256": "d46c3cc427118b37b2738412974c08b412c3d2b0cfb29ff3634ab235b261426d", + "size": 95159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh-result.json": { + "sha256": "775a44ffad890b67be20d057576099881d2b19fc602a4ae38edc6253035edaf5", + "size": 831 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/config.json": { + "sha256": "b4529831a7de33f67351c8cb904c547aeaa6005e2a0a220c2ae44d34fe531e95", + "size": 1541 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/lock.json": { + "sha256": "c9cf59f11d7c26a967eebfa3aee8a41ba42c51eb7a54c5b57001474df56b6d3c", + "size": 2104 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/result.json": { + "sha256": "820ee3a59fb00dc5cb75f06dbde360353e85b4c7a11ad3a1ae39f1c51c18209b", + "size": 36499 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/trial.log": { + "sha256": "bbe49ae9847b03bcc835b6fd70447910a731e117541c5ac67f49b50b86f6ac82", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/attempts/1/ctrf.json": { + "sha256": "2cab2905e5d50aa2a6d5d0713ce02d9d92cda55046347c402c7cb1be29afbfcf", + "size": 2492 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/attempts/1/test-stdout.txt": { + "sha256": "c3d53945b93c1c4d675d491297c1e810beed0eac7468170cb4fc23369b53d3a7", + "size": 1449 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/ctrf.json": { + "sha256": "2cab2905e5d50aa2a6d5d0713ce02d9d92cda55046347c402c7cb1be29afbfcf", + "size": 2492 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/retry-summary.json": { + "sha256": "bfc5efd40036ee3cb99d7c25261057681e5e31b9955b0d4fcf016fbb263573b7", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/test-stdout.txt": { + "sha256": "c3d53945b93c1c4d675d491297c1e810beed0eac7468170cb4fc23369b53d3a7", + "size": 1449 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/result.json": { + "sha256": "e95a5a23811ef62e9d302c85059bee973a23671407ab4ef96f020f92bab0a1c5", + "size": 1142 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/config.json": { + "sha256": "6ef3c739a0cb8a30615d675df06ed1471cd8c9134a19721742ebd2f5003498f3", + "size": 3110 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/harbor.log": { + "sha256": "f83e82fe2baec70dd80e4b6eee5aa58102fbb9789bbef01edb350422961e52e5", + "size": 1281 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/manifest.json": { + "sha256": "333842ab7163ba3ae33c680d07fe4fabcab34fa5a4d34d5f1faa468130ee60c8", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "69ab994ecd3c0b3a7d857179b298e77b4b92c3c66ad136375fa1960c31b43003", + "size": 592 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "4ddc3604b7e0036177ebbc781660b8ba898570ea4bf4dddbccfd2ca1a612187d", + "size": 18711 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "3d2ade665e539a244c6df21c4d4a8150bb364a0f76b1166b4aca1e78a4659f47", + "size": 12019 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "daf73f279a625b69547404177e0f8fdfaf59dfae82b9d69d8adad8126f6c5d45", + "size": 15053 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "869876601c12e3aa782aeda886937b08b793fafb1279810624f16e2986aaee5d", + "size": 18402 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "b9deaca185b5f61be0724d5b50e5997adf42e6bfba190004ecb6608d73ade078", + "size": 14701 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/summary.json": { + "sha256": "64101b431cb2af51aa23b2c1dc0111eba895970c4238606bc1a5df4e7a966df8", + "size": 25055 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/config.json": { + "sha256": "8e4c1b5fd254e707c487184bc38366f5c3dfb9b8dc463de082732ba362b7c0ea", + "size": 1544 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/job.log": { + "sha256": "d33ba5df66979f1b709f536fa81524a90690e64693c17f0dad9788ba8c77b129", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/lock.json": { + "sha256": "1a6853e96705ec60d221d3fa8ceb3aa422f6363e11eab2a65e1e8fdd4341f389", + "size": 3247 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/bootstrap.json": { + "sha256": "8a3916db45911667d182143670fe4edfb134d999147e568a8af01013ff22ca93", + "size": 9047 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/.anonymous-user-id": { + "sha256": "5c60710ceeb5c2a2c992718c430489e11b0b9875f695016258ce3036eaaa0ebc", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/attachments/v1/objects/1a/1a9205d5ae4825c0b40efd3f0de309e03332af1bc3be3324cff6b4282f2a417c": { + "sha256": "1a9205d5ae4825c0b40efd3f0de309e03332af1bc3be3324cff6b4282f2a417c", + "size": 225601 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/cache/attachments/request-images/88/8872a5864fd1fb6ee852a88def52a02933b6de4a77f19a3a742136cdd1fc2247": { + "sha256": "c98eaa3f0ff6ccb6ecaee2d3309630d72c95fbb97d6e63ded434a8c86cfa6fdc", + "size": 106146 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/llm-deepseek/files-v3.json": { + "sha256": "e2787184fe8f70603847ea081b2a3cff3716a1c35eac0bd6f227ed1ef1d9ccac", + "size": 486 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "305c7be770396a0c78dc6051258c736b290b73762ba2868ee3e598a33ffac734", + "size": 462570 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh-events.jsonl": { + "sha256": "8faa97f012c83f1d8770c433b82fd50d068d045e89c7de1a433ce6ab8c6f023d", + "size": 524627 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh-result.json": { + "sha256": "440c91eee9da1fa86c83f31ce9e9efb66e615a3167bc5ae3ecfc4de94b5309d7", + "size": 395 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/config.json": { + "sha256": "c9c41993cc148c1662f1ad50633abbdb7207ca685f839782ac3beb1ac908e891", + "size": 1508 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/lock.json": { + "sha256": "4d90eda4f73f9c0a3e165b39666dc6bc2a4dd2499cc924804aa80feb1e2ccc11", + "size": 2093 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/result.json": { + "sha256": "00ddde593f3721a51d4629de0d322dc56127daafb25ed0b6a9c382b64c4e1a85", + "size": 18410 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/trial.log": { + "sha256": "d33ba5df66979f1b709f536fa81524a90690e64693c17f0dad9788ba8c77b129", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/attempts/1/ctrf.json": { + "sha256": "4a69d55f04cd2d81f5903c9ff491256fbf5519f78f658c3147d0696566eac624", + "size": 2129 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/attempts/1/test-stdout.txt": { + "sha256": "34b5428a6ec54e07c1f7b186fa761f233e5eb6bc79d630fe81aa2b424c6c0617", + "size": 1583 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/ctrf.json": { + "sha256": "4a69d55f04cd2d81f5903c9ff491256fbf5519f78f658c3147d0696566eac624", + "size": 2129 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/retry-summary.json": { + "sha256": "007fc2888cc95d1ff358de21099c658aee73c862a5d3ffb78b0fd8e7a94b8007", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/test-stdout.txt": { + "sha256": "34b5428a6ec54e07c1f7b186fa761f233e5eb6bc79d630fe81aa2b424c6c0617", + "size": 1583 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/result.json": { + "sha256": "cf3efdc136c5b4ec55f493de3ddc39b60ece12356b71b2508d34bb1b7b392eaf", + "size": 1134 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/config.json": { + "sha256": "269089d8ad7d35f433b7a69a64e20cf34685301ad1a6958718344b34536f21d5", + "size": 3088 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/harbor.log": { + "sha256": "1214e585c9566905aa0e9255bfef2159f2ef6c61471e55785652b4411222420b", + "size": 1546 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/manifest.json": { + "sha256": "b854a834dec4f93014e92aab9fb9488c894d463064e9365efb860c6d7401215e", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "f518afe74b19fb3afc53422d64c0a2d66a008ed808a06f943ac6901c526a2487", + "size": 581 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "f14698bb19ce923e66a69928e09f57c15bf8bf2598754a06220ecfb25d952753", + "size": 18700 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "e79fd3f8a177c8995d4f3f952d942f7bae31f499f5bddcefbe5ceecb3aa56f60", + "size": 12008 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "c2d9a1bcf87557772ab3e0b7d4f902d3fda208524f686ae1e285911ceb88ae72", + "size": 15042 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "fcbde320227c0989a846a206d0ddf06b25790e7abcf89f31992775ca8d498b78", + "size": 18391 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "18f161020046c07a0b4d3f76a49cf0e953676373d6eadc11bb61d6cc690c2c9e", + "size": 14690 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/summary.json": { + "sha256": "8478399dae2b8cf9ad050cb32e835c1b9b41909b376d93a0a15755c0fe9e89fd", + "size": 10845 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/config.json": { + "sha256": "5f33d93f2b9a85a5241e45b11b298e220230cb6da8aa0b8f4362bf4f33af452b", + "size": 1542 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/job.log": { + "sha256": "8f9c0c5de8ded83fa536698e90d18113e580bd86cfff1bc01d6e6ac99c9479b5", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/lock.json": { + "sha256": "32eaceeef43cefa611df5474560a5d1e468871f26acef0e17b066d5313a03b39", + "size": 3246 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/bootstrap.json": { + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e", + "size": 15490 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/.anonymous-user-id": { + "sha256": "de1ccf91ca24484fd6c96fddd4171d58cf7254f290f58359ffdb38a880960456", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "e0644370cd6e6c0a43db58461fa1a9d9ed2707c797f1b0ad33b8ffd77f75c2ec", + "size": 73025 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh-events.jsonl": { + "sha256": "f0a93c257e88808cdb0164d14b4b8aae05314a37fc6f4de6413bb1bc69b0fc06", + "size": 83365 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh-result.json": { + "sha256": "2fb06ed034a9e67855fc1bdc993b0de78d90d02d3c0a76e4b27d55dd770cbc9d", + "size": 651 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/config.json": { + "sha256": "99910dbc8e5772c8f614ba2247bd11d20c99d868501c4742e4a1e4af2f2c8e1c", + "size": 1505 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/lock.json": { + "sha256": "4296d5248b6eead253b7ea7212bc0491b6a81c2278b048f75f43d43eeb02830b", + "size": 2092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/result.json": { + "sha256": "1f924d58cb240440f7bf6a579ce51d51558b4aa6361c2cd21fe78578a7c6964c", + "size": 27653 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/trial.log": { + "sha256": "8f9c0c5de8ded83fa536698e90d18113e580bd86cfff1bc01d6e6ac99c9479b5", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/attempts/1/ctrf.json": { + "sha256": "3cd235e86dd07c23bb29d9977f975ca53cc1f65f1fe134f2fb60ee38dbb68d27", + "size": 6513 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/attempts/1/test-stdout.txt": { + "sha256": "1e005e0457f5a1517d55adc255df5b56a8a2feaef76331757269b87888f5a587", + "size": 7087 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/ctrf.json": { + "sha256": "3cd235e86dd07c23bb29d9977f975ca53cc1f65f1fe134f2fb60ee38dbb68d27", + "size": 6513 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/retry-summary.json": { + "sha256": "2b3fec6f55cdf29fcc426051dc46e3c4a03e0fc7608c57f73e55308a0d067248", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/test-stdout.txt": { + "sha256": "1e005e0457f5a1517d55adc255df5b56a8a2feaef76331757269b87888f5a587", + "size": 7087 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/result.json": { + "sha256": "9ebbd685affb67b3c12e21e866385ef211db2c9e6383d58509357ea3a72c83be", + "size": 1130 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/config.json": { + "sha256": "f5effb8af029240b32cf3b1ccda4ab047f63e23e8d8e85befad3eb95b9eecd6a", + "size": 3086 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/harbor.log": { + "sha256": "715dc88404e1db83aae4492ac2895e88af0d444fb5b4e1d00d45c37bd0bc7f93", + "size": 1257 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/manifest.json": { + "sha256": "eece49ade321c217abae0b62394afcc0db75a4099be275bae9a9b64d5e492dd8", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "a0a1cacead633bfdea051d5fad481a38a9717a0e193844841f38a9ab6c76041c", + "size": 580 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "0713ea87e8d5987f82a1571f0c4d86295a8e33dbc7483454f44965ecf2eddbb0", + "size": 18699 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "05f8f3ce5e22c73097715953f2cefa90f12bdda3497d312125e124947d080046", + "size": 12007 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "1352574dbca038f3391300c398a16171829ab9af6917cb3b621c6c8bd54a2d08", + "size": 15041 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "15dd791851e20e0a1efabc0f2840657ceac3517ea08e3130eb63ad7ec864cee6", + "size": 18390 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "7a31a88bfb115934f31552bea2979d74ad0a8e302857c84fc07a7b6cdc9d7974", + "size": 14689 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/summary.json": { + "sha256": "e3d2abbb11e6f8c7a8dfeec37f4635ed193c295b6bdad87fdddb6da1ff104321", + "size": 18083 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/config.json": { + "sha256": "d2df4f2ccbda56842fb5c3d4867d96020bec6157fda5449de0b75a3431db5bf1", + "size": 1554 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/job.log": { + "sha256": "850c0bfa069f204f5aa5a1b2e4972b39de6d71dff117f2453a33d15e4f270175", + "size": 1579 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/lock.json": { + "sha256": "cb9c05125eb40650caf946cce7d8771c74ede6a8c8930efe85b14a7d04ee0ef1", + "size": 3252 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/agent/bootstrap.json": { + "sha256": "fcffcb601c8e963c50a8a668b3f89880eee090e1d90f2219c779013699cc7fa3", + "size": 21474 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/config.json": { + "sha256": "5c508f697704cf1a278e1e6ef092f7c8dab5089c6678ebf8701b1248c05ff745", + "size": 1523 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/exception.txt": { + "sha256": "67cc71bef105517fe1690789812e418d8b27d4aaa36980cb827dcabc55de159a", + "size": 2660 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/lock.json": { + "sha256": "b0391a83974f69d0fb85c2eb2030ea03f263e72deaba470734bccb597e27c6ed", + "size": 2098 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/result.json": { + "sha256": "1e683e794c84c6f143178ce20ddc7ae962bc41452c31cfee27f9863ac20d1b7f", + "size": 7908 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/trial.log": { + "sha256": "789f84bba478edd0eb153b082bcf479ee69a4f012294f1971661ead5b2c00d47", + "size": 1505 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/result.json": { + "sha256": "ce4240899f941c7cff8c1a69a4ffdfef07929e4e4f7be2a8a9da475f49ad9e69", + "size": 1087 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/config.json": { + "sha256": "a02841879dce7c30bc5c91d5019403b808f1882d628fb320c2b85d997bbe162c", + "size": 3098 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/harbor.log": { + "sha256": "14efd5f80bee533f7d380072dded9c5ff2b492b0aeee37bb11480ce5ee3fc451", + "size": 1442 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/manifest.json": { + "sha256": "f0bb5f17b161155815fd7d02a6d9104717166dcb2fd65b259d721cfce80dba71", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "8c64a74a796cbc5276cf6fec93cebae73e6ba94d1ad928ba7239efd2ab7129a6", + "size": 586 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "33011441357cf5ac82fd4b448508d1d4360963703bab2cef41de8ca0fc9d31d9", + "size": 18705 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "4f2de23e64becd4dd8ff72e041b1359ff47373caf210c70fef70ac9113f3dc8a", + "size": 12013 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "3cababee8b6aea5a8110442f45a20f1448f8b50f827fe8a0af012cd448f9deef", + "size": 15047 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "14fe8d3452d5fb0a1abc4a2f788795ab1cec31b4a04c8d5bd521c18f57948bcc", + "size": 18396 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "fd224eb9f2dd2478edf2672b97aeb045ccaebb23d64f97fc4313de76d55aba90", + "size": 14695 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/summary.json": { + "sha256": "5ea9f741ad55baaec08fd88d7e1b2f650efe85b471e3a8421e6e7c2633973a42", + "size": 24419 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/config.json": { + "sha256": "d39e8bef776df6aeb6b14beb6274e9d773948c26a6a16b037ea075028ec76062", + "size": 1564 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/job.log": { + "sha256": "c5ff590ae8374a8330cf8df8b14144d3fe6d2b6fc7c7fdbfcf56df93bc6ad499", + "size": 7637 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/lock.json": { + "sha256": "54763ede2f9ecf1118b8c9e7df6e7e1ac86d045f7bc00c6cb01c1cd3d2fa4b8f", + "size": 3257 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/agent/bootstrap.json": { + "sha256": "569634e2774a66f6a021d30250a11667067e2fb7503586e9cffd6f8eadacdb75", + "size": 21911 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/config.json": { + "sha256": "bac153043e6ce019c24138a9216d93dc1fcb50276b7678b8385e8ec2c9a67112", + "size": 1538 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/exception.txt": { + "sha256": "06df7743d111e65239e3a57cf415f3c566fd3a4ef2c91bf643d787c75505105f", + "size": 5141 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/lock.json": { + "sha256": "421ef7ad9f37170ef20bd7f1425db25edeb85fc54285ce91835e14413a37b5e4", + "size": 2103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/result.json": { + "sha256": "a46c3c8544003a4b9b0a5e7166d38a3c511027b9f47f28e1e698f5ce3b3097ec", + "size": 10024 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/trial.log": { + "sha256": "649070006187e43bed47263196e7e41a2418e42b46968a5321147b7b9d9c0ced", + "size": 7563 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/result.json": { + "sha256": "83e72873c25145206fcf65f4132fdbecbd208080b575d112bc61f2df25528f99", + "size": 1092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/config.json": { + "sha256": "0b08782fc4919bf750aee2e8df7f2fec1ff83259d238a5cd7d4f4758a05d4613", + "size": 3108 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/harbor.log": { + "sha256": "4a3ef901c6403e5c016d0e2303bd9c44a0c083bd38f6500fadd69ee5d8b851cb", + "size": 1456 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/manifest.json": { + "sha256": "dd319d8b72007f38327168b237035984477c465e9c685a05b8773d40ea67bb38", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "e98eb26b40f41fbb61c3e74a381143a9af63fbba781e8758e4976b7b23e516d0", + "size": 591 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "d7e05a52858f7afa6d4045d99f46df0d24628c30e6c182ad52edddea765a3768", + "size": 18710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "f51d390f2bcdcff235fd49513f0909deca0796b731fa0301df0e4bbdcd4e99d6", + "size": 12018 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "2d5eef0c223bda4cfb94b8a163f7c62ed06c3aa01bf9d8196f1fbca059049695", + "size": 15052 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "821453e42e55e78d4d08df1b25b06acbf4e1d99463252feac56f37978926a424", + "size": 18401 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "abc1e0f22ad51a69bd00e0cd5eb1508ba67c507b1ab96655d07d474eadbb038c", + "size": 14700 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/summary.json": { + "sha256": "ad7ec024110bf6d505a48b064aeb994d17b96925b2857927bcd6cd01a7219871", + "size": 24950 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/config.json": { + "sha256": "44d927973b9a25bcc082ebddbcaf2436225572c1596027daf4e52fe23a6c7ec7", + "size": 1550 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/job.log": { + "sha256": "45d52d94ad99e63bcc5f8a0cff967def6acf75768704cc6c517bd1ddc549e120", + "size": 13379 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/lock.json": { + "sha256": "98c8a01418020c74cda0ea32b9d034102ff3cb2219a3fd702311c0a03de1dc6b", + "size": 3250 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/bootstrap.json": { + "sha256": "1e672441c98d4e4e39bbb460a3c2c7e1d185954628be757adf5af4f0eef9a767", + "size": 8485 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/.anonymous-user-id": { + "sha256": "85b79c6b6ec63ec985a5ba94ca1f8d41adcf61a644f60f0e5d3d799f91c5d1f5", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/0e/0ebc72e45e457dafd5962ddf6d96577314312aa2655a4a2a22070d580e39efd8": { + "sha256": "0ebc72e45e457dafd5962ddf6d96577314312aa2655a4a2a22070d580e39efd8", + "size": 22943 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/3e/3eee984a90e4df35db8e8b48928d5e1a83b588b55c50fd9d768833465851679a": { + "sha256": "3eee984a90e4df35db8e8b48928d5e1a83b588b55c50fd9d768833465851679a", + "size": 22832 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/6a/6ae84786f68e98417172a4b7dbc2e175d538e5d99ab543a9d54578aba9bc29e5": { + "sha256": "6ae84786f68e98417172a4b7dbc2e175d538e5d99ab543a9d54578aba9bc29e5", + "size": 26833 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/81/81ebcdf5049cf14069d1c5942c6874ee3f0f5192cfca44309f921320fad9d608": { + "sha256": "81ebcdf5049cf14069d1c5942c6874ee3f0f5192cfca44309f921320fad9d608", + "size": 18261 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/9c/9ced88f6500ead06d77492554505765e3c0965608430d208326e1093bea68966": { + "sha256": "9ced88f6500ead06d77492554505765e3c0965608430d208326e1093bea68966", + "size": 11692 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/ba/ba8cc0c3cb25cf135ab1c4f9d988e35528a04ec9ae893d488e4b5cd4e3528968": { + "sha256": "ba8cc0c3cb25cf135ab1c4f9d988e35528a04ec9ae893d488e4b5cd4e3528968", + "size": 21057 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/ee/ee7d9b185f539550dc7b965d57fc7a67ced9298dfb160f7526c412844dcd47d9": { + "sha256": "ee7d9b185f539550dc7b965d57fc7a67ced9298dfb160f7526c412844dcd47d9", + "size": 32412 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/ef/efe70c559c8779b9b1603422d69aa0875d7c3eed4acfc67b75fc0348fd53328e": { + "sha256": "efe70c559c8779b9b1603422d69aa0875d7c3eed4acfc67b75fc0348fd53328e", + "size": 3813 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/attachments/v1/objects/f4/f40429d67a6738ae5fd877289ff0d536df85c3a3e5333c77f7a552eac3892e26": { + "sha256": "f40429d67a6738ae5fd877289ff0d536df85c3a3e5333c77f7a552eac3892e26", + "size": 24189 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/llm-deepseek/files-v3.json": { + "sha256": "c9e7c68e2336dabfc7a66e1311636fefa57efd5edbce477a1fdc74a628b7777d", + "size": 4012 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "bee845f75f21704a6594a4cdfad6001d85452cfc2676d87dfcba7a474dd39cfc", + "size": 252511 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh-events.jsonl": { + "sha256": "3d74bc6eb5928b841158625a3ea8f248c5203d7b8ebacb92e7535e8da383910f", + "size": 294553 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh-result.json": { + "sha256": "c3b3dfa77c7bb8b5d65eb70c52ac1d6fb93b0675299a316056be60796037402d", + "size": 332 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/config.json": { + "sha256": "50f5320122e7e4eba0a820f747c0632ac98dd0d20dbcb1ecd1dab3d62e16e3fc", + "size": 1517 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/lock.json": { + "sha256": "66d5249a0af8898745e630f286f7f3670c0f94cac03d68604d7d5d41e14dce2b", + "size": 2096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/result.json": { + "sha256": "276a605348ba326675560d046b7d0e5d42596c8c2d58c9ae19b8bae966a8a2b8", + "size": 17773 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/trial.log": { + "sha256": "45d52d94ad99e63bcc5f8a0cff967def6acf75768704cc6c517bd1ddc549e120", + "size": 13379 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/attempts/1/ctrf.json": { + "sha256": "0ac2d266d070ccdfb676491f1d21674561188f9b17b217e8d0d0abc68291d96f", + "size": 740 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/attempts/1/test-stdout.txt": { + "sha256": "fa6dcd755e95aed0c733dcc234986140d131192411d8785160fafe7288887798", + "size": 1173 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/ctrf.json": { + "sha256": "0ac2d266d070ccdfb676491f1d21674561188f9b17b217e8d0d0abc68291d96f", + "size": 740 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/retry-summary.json": { + "sha256": "cf571da57a07ad44dfbb678b31b038e53b9b1709234eb0cb5fc529bcb85c0d18", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/test-stdout.txt": { + "sha256": "fa6dcd755e95aed0c733dcc234986140d131192411d8785160fafe7288887798", + "size": 1173 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/result.json": { + "sha256": "e565ef94bb5c51be9d6ad8f29ad77d2d636422e65b95b24e99a4595a8a6de6c3", + "size": 1136 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/config.json": { + "sha256": "066d9589181e89c1b553064064175377aa73d32f52613bbba680aa2b02695a11", + "size": 3094 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/harbor.log": { + "sha256": "8a38c19efe5849925b8e8059ad7ef1b3918ae4c7509674e5afea210d63c27c79", + "size": 1264 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/manifest.json": { + "sha256": "d296f48ab339d781339af5fbe6fa9250b496b7d1c0d9636876fb27924cc276cf", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "0cd23e8b3c202118dae41ecd7718cec3cbabadc71b1bc4c44b83cd9d9e15a5cb", + "size": 584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "6b2a2f3d6d2c2b69aca3734836ad8b80b8a3a4e512086aba133ab62689a28e78", + "size": 18703 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "6fbe2fa7479ac2c1806fc09683fd92578e354a0029f9da7a1b04dc0488daa630", + "size": 12011 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "9a8e990ba3870969d21209dde044850207eba04e6a10d3b927a229a1fc774450", + "size": 15045 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "c583f527c996accfaa80b7753a2a0b5343fabb198c37ce5ed7d7f2b0dc5b2d8f", + "size": 18394 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "700f4aaa906bad3371a9f05248621175d1e1147414e6917b951f98754fd8210f", + "size": 14693 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/summary.json": { + "sha256": "48235b3e4a6b88691c371a2c70eb0254aaed3dba67f7426b5d2326ca9cc6c9f0", + "size": 10271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/config.json": { + "sha256": "8e4ec073ec50cea54f3b923275ee2e2c564dfa041986b45512a46a600099995f", + "size": 1548 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/job.log": { + "sha256": "9333e639da851a6fa7ac27738166a50a0cee668fbbee9aed5909ef93053aac1b", + "size": 12799 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/lock.json": { + "sha256": "277c33fa8a1748aff1c7730c054116f3d5a775f710840be8691ace78f37cbe95", + "size": 3249 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/bootstrap.json": { + "sha256": "89006797609af114ee0903b557c5ff434b3ef6071d8346100829e653fceb38ca", + "size": 15103 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/.anonymous-user-id": { + "sha256": "364c5591480dbcd8d494710469422a51c5756e5bc2ef0f3602796a2c30231216", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "2f5c3221f07bf28cecff4997e0fe8ab00001ab3ea259885698315f972434da8c", + "size": 39551 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh-events.jsonl": { + "sha256": "7e25269008a09aec00c241197a92ccc2b282732e350fe6c99f00d702c7f5850d", + "size": 46307 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh-result.json": { + "sha256": "d9e17d8e9f5c60d0e263c11eec663a633550ece371f705b8f2b3f91148f1677f", + "size": 76 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/process-supervision.json": { + "sha256": "eb8deb4559f5e566c55691ae383fa0249d537488bc446e7029e76b653396c566", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/config.json": { + "sha256": "8d8c55a3655b05aca76f55f5bae8cd3029244e77bd16e34eecb7eac769433ee1", + "size": 1514 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/exception.txt": { + "sha256": "fad5e5942af2769ba944129d9ebe76945b913d2e4996367ec211ee1050096882", + "size": 7685 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/lock.json": { + "sha256": "2a2537c94edfb51442e456e1ab46963a7fb90dd9f0f00bd134825f3c76fdd1a7", + "size": 2095 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/result.json": { + "sha256": "214469bb2e966535105bb1478124f93dbb16186218ae7fa61feb3ad7a2289590", + "size": 41307 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/trial.log": { + "sha256": "da8bcaf1adbe51c981a0d61f53f43d04addbb70740f25dc176f1ca6a9e1251e5", + "size": 12725 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/attempts/1/ctrf.json": { + "sha256": "12e551d5e256eb47fbb8e2a8082fdfd47e25adad76c96df29a3c75f22a5dedaf", + "size": 5775 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/attempts/1/test-stdout.txt": { + "sha256": "61192bc4a8d7361a021abac1cea705206e9db281c3970f0a8057009bd65a66dd", + "size": 1301857 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/ctrf.json": { + "sha256": "12e551d5e256eb47fbb8e2a8082fdfd47e25adad76c96df29a3c75f22a5dedaf", + "size": 5775 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/retry-summary.json": { + "sha256": "81ba64206851520023c8fa09a84a9d520e33f3fddbff0d49029bbe411f093db2", + "size": 271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/test-stdout.txt": { + "sha256": "61192bc4a8d7361a021abac1cea705206e9db281c3970f0a8057009bd65a66dd", + "size": 1301857 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/result.json": { + "sha256": "e5c76b849573ac94372d1f26621efae5203f84fa562ee58346616390c383861b", + "size": 1270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/config.json": { + "sha256": "2f89479bcaf4717dc24286cf5b6458c5beec5c5f9f23843f4770080ee8dc50ab", + "size": 3092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/harbor.log": { + "sha256": "a6476b2786c8939fc76b2443ddab323c146b4545834d4b81eaebe286e877603b", + "size": 1688 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/manifest.json": { + "sha256": "51ff0c173afe9ee66c5b7d0dcfa0319a0f5703b302f3604bbd091714e80f1f3b", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "6b523e11f6273ac8d69c6fa161f98c93342a8866e05a9fb9ffb240521a54c2b5", + "size": 583 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "1a34f352645870b784051ba38d783938555bbf2f21ead26e151a09c382e18bcb", + "size": 18702 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "bab689695aae968ee20aff4dad1aff5d93d1ee5cc5e74b8fcd3fab80f5837e1a", + "size": 12010 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "a8351af0f727dcc857f094f0e769c310294eaebd201bcfa65c503a91d6174fc8", + "size": 15044 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "a43880bfec537272147bd04624d9c75bfc2c3147ce1db4a49f629fb64c88986f", + "size": 18393 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "0ef39f9aa72eafbc131f51287340e824e881b514b86208dd6374eb3cf9cd0fcb", + "size": 14692 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/summary.json": { + "sha256": "eb6fd2b5c76474f78a6d55a58ae9bf85f53d2c37de476ed58f0e116b88a33636", + "size": 17625 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/config.json": { + "sha256": "3474e5b0c099ec0e2d65bdb2aecf7b00dabcfb6abe156e55f8b09ec81890e6f0", + "size": 1542 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/job.log": { + "sha256": "4cf4038b92512fd76252d59baf09743f3577d9d6016a0394f56dd618c5b02d37", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/lock.json": { + "sha256": "def424d148ea764e17aba9bd62c130aec3c417e913fab767709577b49b5c9227", + "size": 3246 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/.anonymous-user-id": { + "sha256": "d4f08e1cdeb8f0113151486f6d5c84c553590acd60d7700d72e4d03d3af2958c", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "d89a9c99a02c5f3b225286a7baaf1cb7aea1292d19de410e5fc2d47f1fcf09fd", + "size": 1625101 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh-events.jsonl": { + "sha256": "f96e82c2d006c2b6c5855adbf49691f3e4ee3b1f11a11b488762049d1835edb4", + "size": 1832089 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh-result.json": { + "sha256": "364b235a43611d19f4044d40813f5572e3ec663cd7e6ca458000f8eefa044f4b", + "size": 337 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/config.json": { + "sha256": "b31aad87f2a09a9eccc618ec130ec0652ea34bea554b9c39810ac18b47af5351", + "size": 1505 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/lock.json": { + "sha256": "52b7dd65683cd7f1efa811ddb55c67e6141da140717711d8c9016085fd15a63d", + "size": 2092 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/result.json": { + "sha256": "6e7a80996866db583f4b08991823055f2f0971dd26c7b942686c918a00260dae", + "size": 36409 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/trial.log": { + "sha256": "4cf4038b92512fd76252d59baf09743f3577d9d6016a0394f56dd618c5b02d37", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/attempts/1/ctrf.json": { + "sha256": "2912ba7314bd171e47cf05a811f0a0a65f5fbde59644c2c5ae1fbb214ed998f2", + "size": 1775 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/attempts/1/test-stdout.txt": { + "sha256": "b4409c4e80cd3272c7d52ac37e04e70b8ac10040361baceeeb2665a3fcbb90eb", + "size": 17971 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/ctrf.json": { + "sha256": "2912ba7314bd171e47cf05a811f0a0a65f5fbde59644c2c5ae1fbb214ed998f2", + "size": 1775 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/retry-summary.json": { + "sha256": "df85bc4277f0973448621ef8d45897f83614a5c430b2827f2bbd4a82f59dd92d", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/test-stdout.txt": { + "sha256": "b4409c4e80cd3272c7d52ac37e04e70b8ac10040361baceeeb2665a3fcbb90eb", + "size": 17971 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/result.json": { + "sha256": "c4dd4fcf87d4e615e9c143d5b139defcede66c5961a16dc04eb584ea53f4bd04", + "size": 1135 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/config.json": { + "sha256": "96614fdb59d09909d7343c3966868c5663476f23f9319da85b3c59c596935ef9", + "size": 3086 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/harbor.log": { + "sha256": "2a9bf4c750f8c9d0872267a8230660e0632dbd04d275668beb2379d86e156fcf", + "size": 1258 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/manifest.json": { + "sha256": "0f89d986e9251cc5815ed577f9e67de05c9b19136a9a54690f7798a6b6764426", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "05a3ac5d94050c1be7f571ebed44765a257e8e5f62079a8e786b3b41c43e75f1", + "size": 580 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "ea4d37b8ba55c59b38850ce59924c20663064d372034b29add53c0139c14c07f", + "size": 18699 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "c44c59fa8e3c7b34c43cbb4be54770039bf0c00e2bb4872072bf3da5034811e5", + "size": 12007 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "133da55b367afb63acb75166e843a87864c8b90b24e0e0cf664bf5b5b3bcdb78", + "size": 15041 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "3b818cc35f487f5532d4e37227a073477ac045c39c6609cec2bcc026a3de1103", + "size": 18390 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "1a9431dc2841c971c5d7cb7f34db6482f2c0b0664ba2affd5638ec0ece0911b7", + "size": 14689 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/summary.json": { + "sha256": "26eb601e1d8fe9605fac6fe7c62b5b030897bbcf0fc0d6cecd2f28b0b8b61584", + "size": 25032 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/config.json": { + "sha256": "de2c925fd7d0ffab9a9cf2875fcb578902bd8a8396bb1ae8831349562157626f", + "size": 1538 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/job.log": { + "sha256": "ebd6045c1e3f71ce87f284504b3972747983c132ead5b7d1baff9446d12a376b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/lock.json": { + "sha256": "dfa3a294cbc7830aac8f7c9403ad422a46ca1b8b824bd4c28137ff0ffac95772", + "size": 3244 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/.anonymous-user-id": { + "sha256": "770a9cb55aa71263d421e5611bd3fbb7225afc38f0d0da45ee15d135096b834e", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "14d26cb97873cc1c90cb5ea52d3e56fa112a755f765963897401bf9d0cac4c7e", + "size": 227736 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh-events.jsonl": { + "sha256": "04211edfaba0b560f6731376296ca7b6b561005f22bb0af58b0ccf9f20b7261c", + "size": 258600 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh-result.json": { + "sha256": "e070826bbb670263ee32bb756ca90623e38ac5b60b66402a17d681e701b9f013", + "size": 348 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/config.json": { + "sha256": "acfeab64108b5e010972059289044fea31f09fa655d9d0cb17c548622fd5c7fd", + "size": 1499 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/lock.json": { + "sha256": "a20af3b42ac2b9d0d4f223a79f4a848cfc8a4ec9d8e338357c2aa0d187e7ceb9", + "size": 2090 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/result.json": { + "sha256": "7429bf2afe6a63ee8d962706b34c2d83dd397193d5d51e553d94ec8192b638c3", + "size": 38383 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/trial.log": { + "sha256": "ebd6045c1e3f71ce87f284504b3972747983c132ead5b7d1baff9446d12a376b", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/attempts/1/ctrf.json": { + "sha256": "2ad39293cb8b49f009c26b25c92227d3fbdf983c7d70042c6a6614cdd3f290a6", + "size": 756 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/attempts/1/test-stdout.txt": { + "sha256": "b75cc5ad308b4c261c27f0b693dd97a8d6a862af66036b5785a28b87f354a2ec", + "size": 1274 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/ctrf.json": { + "sha256": "2ad39293cb8b49f009c26b25c92227d3fbdf983c7d70042c6a6614cdd3f290a6", + "size": 756 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/retry-summary.json": { + "sha256": "2bdc5193ca6685119434dcaa7060b08745575e52d92763a9f67052a83a35cc9e", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/test-stdout.txt": { + "sha256": "b75cc5ad308b4c261c27f0b693dd97a8d6a862af66036b5785a28b87f354a2ec", + "size": 1274 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/result.json": { + "sha256": "ff977f387ef49d54185841931be59a079715750f528dcda8d92b825cc032ded3", + "size": 1129 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/config.json": { + "sha256": "530a7151f87605aff7f395a20357b1286868524a91bac5a0ae55c89410640ad5", + "size": 3082 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/harbor.log": { + "sha256": "42a36f6931cecc8c901540a05c325269e3ea30231b4af4c8a3728b34c99a88b7", + "size": 1581 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/manifest.json": { + "sha256": "ad5558dac4408bcd388675a83a39005ccb88a09fac8eef3ae65423084b74162b", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "2f3c7313aea73071239c7a41f3807fde8d9b1458232ac3a3b3a57eaec041e4b0", + "size": 578 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "50e32992192e776de09c0ffd29d349f186ecc623d8a0bce50da2f49a4666be70", + "size": 18697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "245318d7bceeb330655672939e3398bea50204887514a8f9d8d8f415ed311674", + "size": 12005 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "4b0ff6cab6d64e2b33eef652d1083c25b5bbe651509108ee31970e6e59c2966a", + "size": 15039 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "af0c2ec8686d58d8b48120627f62690d4c6acb2e769d5f34b594c116a3028ade", + "size": 18388 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "55a2a1dee54c2651619ae0c7d9e9ac57aac733d75980a09d39672138fdd0648d", + "size": 14687 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/summary.json": { + "sha256": "46ff1bf37e688919054a7d4942f2aa4776cae9ad3a5b96dfe66de2a3ff7bf093", + "size": 26704 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/config.json": { + "sha256": "2a583436fd380c4434dcd657641f53905a05a1786ef7c14b931e3e2eb13202e7", + "size": 1576 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/job.log": { + "sha256": "37737405f5cee59cb06a2ade188e65d907c3f1c688df718d1595622258664147", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/lock.json": { + "sha256": "10cee917423abb3e1afaa8a02696befeab2493a60ea3cef2d010c785a7240bbe", + "size": 3263 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/result.json": { + "sha256": "39dda98aabd27196ef973dbe2be8835df2ad7f07cd83ae041a6147456c1e22d8", + "size": 1152 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/bootstrap.json": { + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86", + "size": 21724 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/.anonymous-user-id": { + "sha256": "8971f7cd9e0a59c7764c60659f24e765004c9c205834fefed99a034c807910d9", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "246e4643c46e111f3039e13d90d9b5f72f220e56da142d5145c63caff0b88e02", + "size": 883082 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh-events.jsonl": { + "sha256": "f57fe30e205fa70c52674163266c222ee11fd7718ca564abf0b1a563738df0af", + "size": 989017 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh-result.json": { + "sha256": "b009afe641f46a0acfc93a110d9f53acc49925f5e4eb58a01fe9f91f12cae441", + "size": 527 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/config.json": { + "sha256": "53ec47ae2198d44bba3b3f2accf23d1ad11a622e6bffdbbc0663cc529e0403fb", + "size": 1556 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/lock.json": { + "sha256": "7871dc1a808e67c291381c6b73528f460962f023069b0a8130499236d1edea6c", + "size": 2109 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/result.json": { + "sha256": "ce6dcb948972f30008bc84bf3f25bc86fac08f0a4d369210ee89327a7f9339c9", + "size": 36545 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/trial.log": { + "sha256": "37737405f5cee59cb06a2ade188e65d907c3f1c688df718d1595622258664147", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/attempts/1/ctrf.json": { + "sha256": "ab6b4750c649de9c4ac48d9a8a6305f50441228be2b591659c707117b00128b9", + "size": 739 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/attempts/1/test-stdout.txt": { + "sha256": "4efcd17d33883ef230af8cb5a3ee3c2d2eeabfb3f0e1982dc0a5ca19ce42955f", + "size": 9464 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/ctrf.json": { + "sha256": "ab6b4750c649de9c4ac48d9a8a6305f50441228be2b591659c707117b00128b9", + "size": 739 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/retry-summary.json": { + "sha256": "e41c3df547232ff74d3c6a9955506072a71d29bfdbaebbf6318abac534a13ccf", + "size": 270 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/test-stdout.txt": { + "sha256": "4efcd17d33883ef230af8cb5a3ee3c2d2eeabfb3f0e1982dc0a5ca19ce42955f", + "size": 9464 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/config.json": { + "sha256": "55143fc6d17f7320ee39534c2480ef97e1d76cd9d3911763972399b9d516e76e", + "size": 3120 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/harbor.log": { + "sha256": "194d1d4433aca65170416822c1ee55228c676bb64fc8bdb98e728f872baa637d", + "size": 1292 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/manifest.json": { + "sha256": "d6111a6bb4ec04f97228747719efea70040f30e6f05057bec6b142e791187eaf", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "5fcb7f3b244d1e0495c4aa85a1099f95017fee93fb1ef9f716e79caddc3cf17b", + "size": 597 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "11780ac66c469aa6142acbe1984cc2fc9da26cf7363a0349f3883488b227589f", + "size": 18716 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "8993e08d0ddbfe574ed23b5b393771428090a5f857ec5da655c53d263e0453bb", + "size": 12024 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "4a71ae574e50025a2ac1daf4648fe8d16348d42c9afb63486024fb42e86bebe5", + "size": 15058 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "9a6c85f8e3361d7375d9463d8fd9ea7f3337aaf2951c4735e73e1c1309a4e79f", + "size": 18407 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "35981202c29c1222c6e8879411199a5faf383b601282d3327dd55c65cadc58a6", + "size": 14706 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/summary.json": { + "sha256": "ec83d1073c105b5e0537d1f0e9b9ed792946b3411b5e252f66f98e430bfc730c", + "size": 25066 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/config.json": { + "sha256": "e51268340ffbe3abc1fe51fc0b826976b52ca251877da8d974f1d4d9f27f3377", + "size": 1572 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/job.log": { + "sha256": "0a229a2e1996d29bdf51c11e05bc4a8fbad90c7bad3debf526f3b070476b440d", + "size": 12897 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/lock.json": { + "sha256": "865741f71664183204397d773080114ef920f1df4bf307bb98dd4a315639690b", + "size": 3261 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/result.json": { + "sha256": "1dd5fff03056ecb867ca10557c82e1154d2a6287c2ba05fb62b85d4b6023bfc7", + "size": 1292 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/.anonymous-user-id": { + "sha256": "0dc74a69e1335a246659f68a3b5c19bac5c6173503e91d5a528052b11eeaba86", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "13e014cc1a339a9f8c16cde34aee002223ad0a1f9060c3648f0302c3f3e88738", + "size": 709696 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh-events.jsonl": { + "sha256": "4ad4aca4726fb1a60989217de15d5c3ddc50de7fddcd0ba2f781e202a9d4c306", + "size": 794560 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/process-supervision.json": { + "sha256": "234c3369ed928b4691b85a2261e195dd42f2fcf5556a6cb292ace22db8b2c109", + "size": 75 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/config.json": { + "sha256": "9ac197e3902471b0dc23fbbee8ecdd1b7366277b147097cf50eddd58ec6fe8d9", + "size": 1550 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/exception.txt": { + "sha256": "3db31c2d14b585a222dfe95002d19c8d715c4248f17b41224a6e95a9acc936a8", + "size": 5059 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/lock.json": { + "sha256": "01b4ec673e7eb586a204b02269eedbaf52e567fa6f38c7eadd8d48e73044d322", + "size": 2107 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/result.json": { + "sha256": "bbd946c59c4aea9ed60f6233def0fa4d97ce45b77650b12798fa01ceaeb6ea56", + "size": 43971 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/trial.log": { + "sha256": "7d5613e1d7303b60d682642d18c6cdde1228da9f4fd71c2ab1d6541fa182a88b", + "size": 12710 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/attempts/1/test-stdout.txt": { + "sha256": "d92c853dd2b021a6c77661b03c422071d2bb1bffcd4db8fd5a0a44e3c20cc5ce", + "size": 1883 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/retry-summary.json": { + "sha256": "c639c755a9d5d93f321985096563a85955f8293dc98a428fe59a0bc8e8f1231f", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/test-stdout.txt": { + "sha256": "d92c853dd2b021a6c77661b03c422071d2bb1bffcd4db8fd5a0a44e3c20cc5ce", + "size": 1883 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/config.json": { + "sha256": "0cda3c9547114934803d112b09228b6f415ecc73cb28d02a4668149a24536b46", + "size": 3116 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/harbor.log": { + "sha256": "4612b40c58b2519056c4c3032a95dbf64545f3c991d229a634735ecc5befcd94", + "size": 1625 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/manifest.json": { + "sha256": "88a5be96e11de4d421b3751d836776344b26ba21e8297832fd3369b1f6debd9d", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "d684a4509d3848d38be9699d9b30ee5e43b995e168914092d479091c7428de95", + "size": 595 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "091b4458aa7e9ff1bd9e365f168a5aac817653dc47e98c7de7998ff57815dbb1", + "size": 18714 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "3ca13bca117b562431c663475e3579ccd4af93a368956f2a7f8e23d64ce0fd77", + "size": 12022 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "ea573da41552e18fc09628e0488dc87c639aec6e1a86213cf758459b1e47522c", + "size": 15056 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "3f6ee2f8b145039ac63e3d877bc495bbe679d3cdb349be092904d3705997eca8", + "size": 18405 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "35bbfc31c9848ed630dcb4b13c871829c9121a8d36e3fdaee9c7e5bafff0e953", + "size": 14704 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/summary.json": { + "sha256": "b12c2487c5c653b7b26e4416f55370894c93b348ffe84711419f08277aea95ee", + "size": 26735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/config.json": { + "sha256": "1509953e67d30d184780a094e1f115e5459ce09384f654321b32b0d6d030f40a", + "size": 1568 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/job.log": { + "sha256": "ae88953072e2d4ea99824bb29d4e74512900ecd25bb82982e50fc8515d3a6bcf", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/lock.json": { + "sha256": "4d84f85538d83b4447ec2e77e69cb7fd7e5fd79d354b682c8c054912249f3f2b", + "size": 3259 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/result.json": { + "sha256": "d67a3d34ecba2b47cba0ac53a2d7a22d05b0c8dc1d22ee7f3c16a71b79d137cb", + "size": 1144 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/bootstrap.json": { + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809", + "size": 23275 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/.anonymous-user-id": { + "sha256": "6c73209c5e482e21cb3cc9b854d03a25935252ca7e7a4df44730624b10c4db88", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "3ce59b16bd7a80313589c9297a306685de35fe3b72e75770628eb278c3ce9005", + "size": 374769 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh-events.jsonl": { + "sha256": "4619ce1d5a34be369ba4aa4b3d9bbe54472096aee99dfcdadcb53c92f5f4d6da", + "size": 415967 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh-result.json": { + "sha256": "082539d153b799604aadd41eedd3d4686cf5af202f5e86170214f860b372c8d6", + "size": 686 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/config.json": { + "sha256": "3c414528033a07365c484ab9381c6d07099001a8146db28561ea9122ad0f4ee7", + "size": 1544 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/lock.json": { + "sha256": "d187ad870c50587c818f491e975cea5e4c782e0440516ac8aa26f318aebdc232", + "size": 2105 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/result.json": { + "sha256": "d06d9acad5f9df54a7430dedd63542e88f240dedebf0517b19980adb22416e39", + "size": 38503 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/trial.log": { + "sha256": "ae88953072e2d4ea99824bb29d4e74512900ecd25bb82982e50fc8515d3a6bcf", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/attempts/1/test-stdout.txt": { + "sha256": "862db79cb138fe5898e2967cf92dbaf6627139d770e5e1271e7923a47f094d50", + "size": 1503 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/attempts/2/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/attempts/2/test-stdout.txt": { + "sha256": "001335f99efb1ac4511114fd5274d30802fbf47e78a69922f8700dcdcd742c8e", + "size": 1282 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/retry-summary.json": { + "sha256": "c93db7d2a7b90856b956a7c8704b7f02b380503c5a3b5feca67b0e81741045f6", + "size": 373 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/test-stdout.txt": { + "sha256": "001335f99efb1ac4511114fd5274d30802fbf47e78a69922f8700dcdcd742c8e", + "size": 1282 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/config.json": { + "sha256": "7b136d0a86d5cb62dd83805a320a76f288880f27d927f81ab0cfeffe6c2b4a6a", + "size": 3112 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/harbor.log": { + "sha256": "1382446b958d047eb2ae6ba9ae71f3f80f54b746f66881335dd855c634e3b27a", + "size": 1284 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/manifest.json": { + "sha256": "b073682154393d82f6b6de279ac58f2fbb4eb9e39f9ff6967bd5b3fd89980e3c", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "87f5f727ba2595f7fcbbd070021eeffff98027098c92eb6da53c3b3834ff168d", + "size": 593 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "2b64b5f98a6bc59b7025d102f6cb0d53006833555964cf4e93f7422ac93fb1a4", + "size": 18712 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "05b0278b3e6337fba9d4d83eb61de6bcec8ba01e2d3d64e163526e3c3781eb8b", + "size": 12020 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "03c21f635b92dadf2372fa4dae12532145dca654e18105d05f30a0cb531d279f", + "size": 15054 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "4aadafd28f0676e38c69d8e2fae1902218aefb29c566d4899d6973d99407261e", + "size": 18403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "3c46ba7bbf4bc9ec9fa1fd0411fc094c04f505386bbd5efa0f8ed61a83fa2da8", + "size": 14702 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/summary.json": { + "sha256": "ea7aa7db1c7b759754d9929f24021d9378d4bd7ecf5de3520c42b6ee2421a5bb", + "size": 26868 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/config.json": { + "sha256": "40eca51f9498cc54f0b794ec0248102820312006dc8ffc879fffb0a1bca6e57a", + "size": 1552 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/job.log": { + "sha256": "557d2b559a85c4210543051765ea61d74e37c605f03faac1b40e371446384c77", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/lock.json": { + "sha256": "b304a8ea0f28d91711b5b2303e4da4de27997af4c798247d7cdf5613147303db", + "size": 3251 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/result.json": { + "sha256": "39671ee6d505b5e03d257a527cbec7589985b46e75ef8d9dbc9f65f681dc29a1", + "size": 1138 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/bootstrap.json": { + "sha256": "106fca350e150fcc87b6003fc699eec6e057797103b83a4c169cf5c14a6b293c", + "size": 19374 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/.anonymous-user-id": { + "sha256": "fb5189e8a5b6e9c293aab340d58dbe8d17fc3437379ce391f9f6b09ec2bf1b49", + "size": 37 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "b1c84f0065ccf46d834fea4a18d5ec27ac512501de37ab98e65ec142b813d6ec", + "size": 335472 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh-events.jsonl": { + "sha256": "70014d4db3875f0d0a369c10e47aa09d7654c21de1e4f5be407c8f118227bbd9", + "size": 379320 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh-result.json": { + "sha256": "066e72c7a3c33c09619604791636111472388d849d2ced1ad4d86f205517396d", + "size": 271 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/config.json": { + "sha256": "4386b351fd1f090e7da76ff5040a01004d5e21be12113206c5b0b54db193d520", + "size": 1520 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/lock.json": { + "sha256": "7400beb2d66705e35c30a513cd6e0f03258bf444c1cb0862efe5c5694d3a8d8a", + "size": 2097 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/result.json": { + "sha256": "9282ab7ec0f5b980db603031a78dddd20add13400918c58d720e8d05f68f1222", + "size": 32913 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/trial.log": { + "sha256": "557d2b559a85c4210543051765ea61d74e37c605f03faac1b40e371446384c77", + "size": 12735 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/attempts/1/ctrf.json": { + "sha256": "cd06eb4d3084b68383e68e5d5d9003d0be981ce72e7ce9b7bf6f58d2bb9bc101", + "size": 1466 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/attempts/1/test-stdout.txt": { + "sha256": "166f15fa270868102a648fd0a846285db328a9c91964451a36c4fed8937c89ce", + "size": 1403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/ctrf.json": { + "sha256": "cd06eb4d3084b68383e68e5d5d9003d0be981ce72e7ce9b7bf6f58d2bb9bc101", + "size": 1466 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/retry-summary.json": { + "sha256": "1617f955d7afcd632659daf0e6109b63b73ab0689e175fa10e9ffec7bde45e8c", + "size": 269 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/test-stdout.txt": { + "sha256": "166f15fa270868102a648fd0a846285db328a9c91964451a36c4fed8937c89ce", + "size": 1403 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/config.json": { + "sha256": "5bffbeeea5a40725c5bf95c87093eaadcc52b59e9d555472a55e95ab52a17d66", + "size": 3096 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/harbor.log": { + "sha256": "1ba4c07a943c0e9a0e55dc8baf7d3d16d8dc6a4a825ca09a834666e22ad1f117", + "size": 1267 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/manifest.json": { + "sha256": "1422633177e16f08b3176abac96fb3c40ca252f48668a6b1780b9d39583a1c68", + "size": 3669 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "992abeab73b4760b88c8a9ffc3eaac6a5270631c90986e8892006ebe4cb9b8e7", + "size": 585 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "915fc3206b0d477e842c3301780727ce5dcbafeb39a8eb8ea9d070de2a840df1", + "size": 18704 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "2d9bc42b78e2adcb59036492733e43c39411feaa4f8672eb6c1df509e455b2dd", + "size": 12012 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "39e5e0128cf57e246b33d2cdd27f63472a522ed2d7954a75f8ffe99b0d037871", + "size": 15046 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "fd1943d20351cb6dc80d3863fdfafcd1c1206c040759559e4cfc4a7a18a2f836", + "size": 18395 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "fcd2d79cec2ec7d6d491a5cfeecd203c14a8bc9f3f2a305246e41f778e209a98", + "size": 14694 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/bootstrap.py": { + "sha256": "7e79712812e3554a4f3912448c095f5966f59b35f4e5dca285dc186487956a13", + "size": 7892 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/summary.json": { + "sha256": "cbbf572b7a201178ee4849443a38a96ad2bb4c673055f67b169ea5ed483de80d", + "size": 22355 + }, + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/dsh-freeze-outputs.py": { + "sha256": "9f038fbe131ad3a01ce05742b03a3565f6e330e48b80cd714d30d7ce15f523d3", + "size": 2241 + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json new file mode 100644 index 0000000..7a4673e --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json @@ -0,0 +1,18 @@ +{ + "checked_at": "2026-10-31T15:50:39.815342+00:00", + "profile": "sdk-minimal", + "finish_reason": "completed", + "local_mock_requests": 2, + "real_model_calls": 0, + "server_reachable_before_sdk_close": true, + "server_reachable_after_sdk_close": false, + "probe_sha256": "8dc0d6fe2f627f472e8db8b4e1a6f3aa937d8d5b96d2235197e02db62857c4f5", + "upstream_fixture_sha256": "c52ace60539a52721fcdb5b61f7d8f2b9806dbbba9cb2f5fcd2d3b2541b61be9", + "runtime_sha256": "28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2", + "schema_version": 1, + "phase": "post_freeze_diagnostic", + "upstream_source_ref": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "script": "benchmarks/harbor/scripts/probe_deepseek_lifecycle.py", + "command": "jobs/dsh-harbor-venv/bin/python benchmarks/harbor/scripts/probe_deepseek_lifecycle.py --upstream-source jobs/deepseek-harness-build-f3803c9/source --runtime jobs/deepseek-harness-build-f3803c9/source/dist-exe/deepseek-harness-sdk-runtime-linux-x64", + "interpretation": "With this pinned upstream SDK/runtime, a nohup HTTP server launched through bash is reachable before normal SDK context closure and unreachable afterwards. No benchmark task or model output was replayed. Raw task scores remain unchanged." +} diff --git a/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md b/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md new file mode 100644 index 0000000..72f0a3c --- /dev/null +++ b/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md @@ -0,0 +1,281 @@ +# Official DeepSeek Harness on Pi's 34-task development pool + +All **34 registered jobs finished on October 31, 2026**: **22 official passes, +six scored failures, and six unscored errors**. There were 32 model executions; +two jobs failed during setup. Two scored failures have a confirmed SDK service +lifecycle caveat. These results describe the official Harness's `sdk-minimal` +configuration through its Python SDK, not every Harness interface or a complete +34-task pass rate. + +On the **26 tasks with accepted scores in both runs**, this configuration passed +**21**, versus **22** for [Pi](pi-deepseek-flash-dev34-20261024.md). +`dna-assembly` changed from failure to pass; `kv-store-grpc` and `pypi-server` +changed from pass to failure. Both regressions involve services available during +agent self-tests but absent during verification. A separate keyless probe +confirmed that normal SDK closure stops a background server. This comparison +does not establish a causal difference in model or harness capability. + +The [audited results](../../results/deepseek-harness-dev34-20261030.json) retain +all 34 trial IDs, raw and analytical outcomes, official test counts, timing, +usage, diagnoses, and evidence hashes. The +[registration](../../configs/deepseek-harness-dev34-20261030/experiment.json) +and [configuration history](../../configs/deepseek-harness-dev34-20261030/README.md) +preserve preparation and execution decisions. All outputs were frozen at +**2026-10-31 15:47:11.582274 UTC**, before new solution trajectories or hidden +test contents were reviewed. The +[freeze manifest](../../configs/deepseek-harness-dev34-20261030/outputs-freeze.json) +covers **2,506 files**, all rehashed during audit. Raw traces remain local. + +## Configuration and preserved protocol + +| Property | Value | +| --- | --- | +| Pool | Exact existing Pi selection of 34 development tasks, unchanged revisions, task files, images, resources, and official tests; the other 55 tasks remained unopened | +| Agent | Official `deepseek-ai/deepseek-harness`, version `0.2.2-alpha.2`, source `f3803c96974ad4a94195fb97e5c637e9c2a3afe3` | +| Profile | Upstream benchmark-recommended `sdk-minimal`, driven through the public Python SDK | +| Behavior | Native system prompt, persistent bash, image reading, working-directory tool, compaction, request retries, and generation defaults; fresh home/session per task | +| Isolation | No personal instructions, skills, extensions, settings, or prior sessions; task tools execute inside the task container | +| Model | `deepseek/deepseek-flash`, `high` reasoning, official Messages endpoint `https://api.deepseek.com/anthropic` | +| Generation | Temperature and SDK `max_tokens` unset; upstream output ceiling 256,000 | +| Adapter | Frozen source `5b503ab6d04af2f6a719ec0dbe5418bfad4bf128` in `jobs/deepseek-harness-runtime-source` | +| Limits | Native task agent/verifier timeouts and resources; setup limit 1,800 seconds | +| Queue | Same Pi order; concurrency two with consecutive pair barriers; no whole-trial retries | +| Verification | Original scripts and per-attempt limits; only verifier timeouts may retry, at most three total attempts on the same agent output | + +The official [agent integration guide](https://api-docs.deepseek.com/quick_start/agent_integrations/) +checked October 30 and 31 maps `deepseek-flash` to DeepSeek-V4.1-Flash-1017; +the [alias record](../../configs/deepseek-harness-dev34-20261030/model-alias-check.json) +records the later check. This remains a mutable alias, not proof of an identical +backend checkpoint across dates. Pi used its native OpenAI-compatible endpoint +and a 384,000-token ceiling. Neither run imposed Nano's 100-reply limit or +finalization reminders. These endpoint, budget, date, and lifecycle differences +make this a comparison of complete configurations. + +The user's original upstream checkout remained clean and unchanged. An isolated +copy was built with the upstream frozen pnpm lockfile (Node 26.7.0, pnpm 11.28.5). +All eight artifact-build gates and the official keyless SDK smoke passed. +The offline distribution contains the SDK, native runtime, and five dependency +wheels. Source bytes, distribution bytes and file modes are pinned. + +```text +plan SHA256: fb59d5618c44e691babb80b2ea8323eb45edf9f45383e61344d747b870bc6d6e +freeze SHA256: cc421c1429450ac41d355d0df8333f4ab10ca4da0bf47f24bc3892d0724d9c24 +results SHA256: e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396 +selection SHA256: 4572b1cc432a16b9c5d1dd554c8d2189ae80414f386a90bf03829bc4d0c3da85 +source archive SHA256: 0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4 +distribution SHA256: d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6 +runtime SHA256: 28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2 +lifecycle diagnostic SHA256: 4257a96a3415a7d389fb7b3fb10f819341714ba833b0911eb2a6468171e70c47 +``` + +## Installation, execution boundaries, and error accounting + +All **34 installation-only checks passed before the first model job**, including +SDK initialization and Linux supervisor probes. QEMU's `qemu-alpine-ssh-v1` +and model recovery's `pytorch-model-recovery-v1` dependency profiles passed; +the other **32 profiles are explicitly `not_configured`**. The +[pre-model metadata amendment](../../configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json) +corrected a historical description of the profile count without changing any +execution setting. The [preflight evidence](../../configs/deepseek-harness-dev34-20261030/preflight-evidence.json) +does not guarantee that later network-dependent setup will succeed. + +The first WAL installation smoke failed because the Python `dsh` launcher needs +an explicit `DSH_HOME` even for its version query. The adapter supplied an +isolated version-probe home before registration; the corrected smoke passed. +Both records remain in [setup evidence](../../configs/deepseek-harness-dev34-20261030/setup-evidence.json), +and neither is a scored model attempt. Validation included 280 Harbor tests +before that integration fix and 11 DeepSeek-focused tests after it. The final +full Harbor suite also passed all 280 tests. Its first invocation passed 273 +and had seven local HTTP fixtures blocked by sandbox socket permissions; the +complete rerun with local sockets permitted passed in 17.56 seconds. + +Every one of the **32 model executions** has a confirmed completion or +cancellation boundary before official verification. No recorded SDK event +occurred after agent end or at/after verifier start. The three agent timeouts +were sampler, model extraction, and pipeline parallelism, each with its native +900-second limit; cancellation stopped **10, 4, and 8 processes**, respectively. +Sampler and model extraction have accepted official failures. Pipeline has a +separate verifier dependency error and remains unscored. + +Native rewards remain **22 ones, 10 zeros, and two nulls**. Unchanged standard +summaries contain **28 `scored`, four `execution_error`, and two `setup_failed`** +records. The analytical audit distinguishes official assertions from incomplete +setup, terminal API failure, and verifier errors; it does not rewrite native +exceptions or rewards. + +| Unscored task | Category and evidence | +| --- | --- | +| pytorch-model-cli | Setup: original uv 0.9.11 installer fetch failed with curl 35 / `SSL_ERROR_SYSCALL` at `releases.astral.sh:443`; `NetworkConnectionError`, no model execution | +| pytorch-model-recovery | Setup: native `AgentSetupTimeoutError` after 1,800 seconds in `verifier_dependencies`; SDK/supervisor checks passed but this formal dependency check did not finish; no model execution | +| multi-source-data-merger | Verifier: `pyarrow==22.0.0` download/extraction timed out with `UV_HTTP_TIMEOUT=30`; official pytest never began | +| query-optimize | Execution: SDK `finish_reason=error`, terminal `TRANSPORT` error after six attempts of one provider request, including five native request retries; native `NonZeroAgentExitCodeError`; later official tests found missing SQL output | +| torch-pipeline-parallelism | Verifier: after confirmed agent timeout, download of `nvidia-cusolver-cu12==11.7.1.2` required by `torch==2.7.0` timed out; official pytest never began | +| torch-tensor-parallelism | Verifier: first attempt reached 900 seconds; second failed downloading `torch==2.7.0` with `UV_HTTP_TIMEOUT=30`; official pytest never began | + +There were **three extra verifier attempts**: sampler timed out twice at 900 +seconds, then its third attempt completed in 26.3 seconds; tensor parallelism +timed out once at 900 seconds, then its second attempt ended with the dependency +failure. These used the same outputs and original scripts. Dependency failures +did not trigger retries. No complete task, model output, or setup failure was +replayed. Native request retries within the agent are separate from these +verifier retries and from whole-trial retries. + +## Per-task outcomes + +`—` denotes no analytical score or no execution, as appropriate. Pi values are +its audited scores, not its raw rewards. `†` marks the service lifecycle caveat; +`‡` marks Pi's reference-exposed pipeline pass. + +| Task | Pi audited | Harness raw | Harness audited outcome | Official tests passed / failed | Agent seconds | +| --- | ---: | ---: | --- | ---: | ---: | +| adaptive-rejection-sampler | — | 0 | Timeout; confirmed stop, failed | 4 / 5 | 899.4 | +| build-cython-ext | 1 | 1 | Passed | 11 / 0 | 593.1 | +| cancel-async-tasks | 0 | 0 | Failed | 5 / 1 | 26.3 | +| cobol-modernization | 1 | 1 | Passed | 3 / 0 | 172.4 | +| db-wal-recovery | — | 1 | Passed | 7 / 0 | 34.1 | +| fix-code-vulnerability | 1 | 1 | Passed | 6 / 0 | 42.1 | +| git-multibranch | 1 | 1 | Passed | 1 / 0 | 83.5 | +| kv-store-grpc † | 1 | 0 | Failed | 5 / 2 | 45.8 | +| log-summary-date-ranges | 0 | 0 | Failed | 1 / 1 | 21.0 | +| merge-diff-arc-agi-task | 1 | 1 | Passed | 5 / 0 | 110.5 | +| model-extraction-relu-logits | 0 | 0 | Timeout; confirmed stop, failed | 0 / 1 | 899.4 | +| multi-source-data-merger | 1 | 0 | Unscored: verifier error | — | 27.8 | +| nginx-request-logging | 1 | 1 | Passed | 8 / 0 | 138.0 | +| openssl-selfsigned-cert | 1 | 1 | Passed | 6 / 0 | 36.2 | +| pypi-server † | 1 | 0 | Failed | 0 / 1 | 40.2 | +| pytorch-model-cli | 0 | — | Unscored: setup error | — | — | +| pytorch-model-recovery | 1 | — | Unscored: setup error | — | — | +| qemu-alpine-ssh | 1 | 1 | Passed | 1 / 0 | 270.2 | +| query-optimize | 0 | 0 | Unscored: API error | 2 / 4 | 231.7 | +| regex-log | 1 | 1 | Passed | 1 / 0 | 90.3 | +| torch-pipeline-parallelism ‡ | 1 | 0 | Unscored: verifier error | — | 899.3 | +| torch-tensor-parallelism | 1 | 0 | Unscored: verifier error | — | 131.3 | +| write-compressor | 1 | 1 | Passed | 3 / 0 | 143.6 | +| constraints-scheduling | 1 | 1 | Passed | 3 / 0 | 24.3 | +| custom-memory-heap-crash | 1 | 1 | Passed | 6 / 0 | 131.7 | +| dna-assembly | 0 | 1 | Passed | 1 / 0 | 944.5 | +| llm-inference-batching-scheduler | 1 | 1 | Passed | 6 / 0 | 503.8 | +| path-tracing | 1 | 1 | Passed | 5 / 0 | 229.8 | +| schemelike-metacircular-eval | 1 | 1 | Passed | 1 / 0 | 902.8 | +| caffe-cifar-10 | 1 | 1 | Passed | 6 / 0 | 2200.5 | +| circuit-fibsqrt | 1 | 1 | Passed | 3 / 0 | 461.1 | +| mteb-leaderboard | 1 | 1 | Passed | 2 / 0 | 1172.2 | +| regex-chess | 1 | 1 | Passed | 4 / 0 | 1907.7 | +| build-pov-ray | 1 | 1 | Passed | 3 / 0 | 891.9 | + +Counts come from the final official pytest summary, preserving parameterized +cases; agent self-tests and embedded upstream test runs are not added. The +**28 accepted evaluations** executed **118 cases: 107 passed, 11 failed, zero +errors/skips**. The API-interrupted, unscored query trial separately ran **two +passed and four failed** cases. The other five unscored trials never reached +official pytest. + +| Scored failure | Official assertion or missing artifact | +| --- | --- | +| adaptive-rejection-sampler | `Rscript` and generated sample files missing after the confirmed timeout; installing R when absent is explicitly part of the task, so this is a task failure, not a verifier dependency exclusion | +| cancel-async-tasks | Excess-task cleanup messages: zero, expected two | +| kv-store-grpc † | Port 5328 not listening; connection refused | +| log-summary-date-ranges | Today ERROR count 74, expected 370 | +| model-extraction-relu-logits | `/app/stolen_A1.npy` missing after the confirmed timeout | +| pypi-server † | Installation of `vectorops==0.1.0` from `localhost:8080` failed with the local index unreachable | + +These are official findings, not exhaustive root-cause proofs. + +## SDK service lifecycle diagnostic + +KV's agent successfully exercised `SetVal`/`GetVal` with value 123. PyPI's agent +started a `nohup` server, fetched the index, and installed and exercised the +package in a separate virtual environment. Both services were unavailable to +the official verifier after the SDK context had closed. + +The pinned upstream SDK's normal shutdown disposes managed terminals and their +descendants. The shared Linux supervisor permits solution services after normal +completion, but that cannot preserve services already stopped by the SDK. +The post-freeze [diagnostic](../../configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json) +used the upstream keyless mock fixture and the exact pinned runtime, with **two +local mock requests and zero real model calls**. A `nohup` Python HTTP server +was **reachable before SDK close and unreachable after close**. The +[reusable probe](../../scripts/probe_deepseek_lifecycle.py) records its own, +fixture, and runtime hashes: + +```bash +jobs/dsh-harbor-venv/bin/python benchmarks/harbor/scripts/probe_deepseek_lifecycle.py \ + --upstream-source jobs/deepseek-harness-build-f3803c9/source \ + --runtime jobs/deepseek-harness-build-f3803c9/source/dist-exe/deepseek-harness-sdk-runtime-linux-x64 +``` + +This diagnostic was performed after all benchmark outputs were frozen. It +confirms a lifecycle limitation of the measured SDK integration, not that both +tasks would pass every official assertion under a different lifecycle. Both +zeros remain scored failures of this configuration. No lifecycle fix, server +restart, or regrading was applied to the registered run. + +## Interpretation against Pi + +Pi's audit records **26 passes, six failures, and two unscored legacy timeouts**. +Its sampler boundary was invalid and its WAL boundary unverified; its pipeline +pass accessed public reference tests and a solution. Those qualifications remain +in the [Pi report](pi-deepseek-flash-dev34-20261024.md). + +The 26 common accepted tasks yield **Harness 21 / Pi 22** passes. The only gain +is `dna-assembly`; the two losses are `kv-store-grpc` and `pypi-server`, both +with the lifecycle caveat. `db-wal-recovery` passes here but was unscored for +Pi, so it is not a paired gain. Sampler now has a valid official failure but +also lacks a valid Pi comparison. The six new unscored tasks likewise cannot be +used to infer gains or losses. The common subset is determined by evaluability +and is not an unbiased replacement for the registered 34-task pool. There is +no statistically controlled ranking, complete paired pass rate, or claim that +the observed changes isolate the harness alone. + +## Exposure, usage, timing, and reproducibility + +After freezing outputs, a targeted scan covered **675 recorded tool calls** +and matching candidate outputs for benchmark repositories, hidden paths, and +environment access. No external Terminal-Bench reference retrieval or official +hidden-test access was observed in that scan. A calendar `PRODID` and a +task-provided C++ canary produced benign Terminal-Bench string matches. MTEB +downloaded public `embeddings-benchmark/results` application data at +`3125e8302256ac01d4dd68b6f92902b3e6a78987`; this is distinct from Terminal-Bench +tests or solutions. The development pool was already exposed historically; +the targeted scan does not prove complete absence of exposure or exfiltration. + +An exact-byte scan of all **2,506 frozen files** found the authorized provider +credential in **zero files**. This is limited to that key and these files; +it does not alter the historical Pi credential findings. Raw trajectories and +credential values are not published. The maintainer's October 31 authorization +and model launch are recorded in [launch evidence](../../configs/deepseek-harness-dev34-20261030/launch-evidence.json). + +| Known usage or timing | Value | +| --- | ---: | +| Input tokens, including cached input | 25,052,194 | +| Cached input tokens, already included above | 24,346,112 | +| Output tokens | 634,263 | +| Cost | Unknown | +| Summed agent execution seconds | 14,306.464573 | +| Queue wall seconds, including setup and verification | 16,984.894313 | +| First job start (UTC) | 2026-10-31 11:03:13.156046 | +| Last job finish (UTC) | 2026-10-31 15:46:18.050359 | + +Usage is a **known subtotal**. Sampler, model extraction, query optimization, +pipeline parallelism, DNA assembly, and LLM batching have partial usage. Failed +native retries and interrupted requests can omit usage even when a later +completion succeeds. Do not add cached tokens to input again or infer a bill +from these records. Different network conditions, setup failures, verifier +retries, and accepted task sets prevent a controlled speed or cost comparison. + +Final validation rechecked source/task/configuration hashes, all preflight-before- +model boundaries, every pair barrier, and **68 original image checks** across +preflight and formal jobs. Original package sources and task images were +preserved; cache hits required matching filename, size, and SHA256. Only newly +introduced image references were evicted without force, retaining protected +images. All 34 jobs finished and **zero task containers remained active**. + +The [configuration README](../../configs/deepseek-harness-dev34-20261030/README.md) +contains the build and execution commands. Every job uses the standard +`uv run --project benchmarks/harbor python -m harbor_adapter run --config ...` +entry point from the frozen source. Native evidence is retained under +`jobs/deepseek-harness-runtime-source/jobs/`, with orchestration and image events +under `jobs/deepseek-harness-dev34-20261030-orchestration/`. Published JSON and +hash manifests allow provenance inspection; reproducing from source also +requires the registered task packages, images, built distribution, and provider +access. Existing registered jobs refuse overwrite or replay. diff --git a/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md b/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md new file mode 100644 index 0000000..6c8590d --- /dev/null +++ b/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md @@ -0,0 +1,252 @@ +# DeepSeek 官方 Harness 在 Pi 的 34 道开发题上的结果 + +**34 个注册作业已于 2026 年 10 月 31 日全部结束**:**22 道官方验证通过、 +6 道计分失败、6 道因错误不计分**。其中实际执行模型的有 32 道,另两道在安装 +阶段失败。两道计分失败存在已确认的 SDK 服务生命周期限制。本报告描述通过 +官方 Python SDK 运行的 Harness `sdk-minimal` 配置,不能代表 Harness 的所有 +接口,也不构成完整的 34 道题通过率。 + +在**双方都有有效评分的 26 道题**上,本配置通过 **21 道**, +[Pi](pi-deepseek-flash-dev34-20261024.md) 通过 **22 道**。 +`dna-assembly` 从失败变为通过;`kv-store-grpc` 和 `pypi-server` 从通过变为 +失败。后两题的服务在 agent 自测时可用,在正式验证时已不可用。独立的无密钥 +探针确认,正常关闭 SDK 会停止后台服务。因此,这一对照不能证明模型或 +harness 能力存在因果差异。 + +[审计结果](../../results/deepseek-harness-dev34-20261030.json) 保留全部 34 个 +trial ID、原始与分析结果、官方测试数量、耗时、用量、诊断及证据哈希。 +[实验注册](../../configs/deepseek-harness-dev34-20261030/experiment.json) 和 +[配置历史](../../configs/deepseek-harness-dev34-20261030/README.md) 保留准备与 +执行决策。全部输出于 **2026-10-31 15:47:11.582274 UTC** 冻结,之后才查看 +本次解题轨迹和隐藏测试内容。 +[冻结清单](../../configs/deepseek-harness-dev34-20261030/outputs-freeze.json) +覆盖 **2,506 个文件**,审计时已逐一重新计算哈希。原始轨迹仅保存在本地。 + +## 配置与保留的实验协议 + +| 属性 | 值 | +| --- | --- | +| 题池 | Pi 已使用的同一组 34 道开发题,题目版本、文件、镜像、资源和官方测试均不变;其余 55 道题保持未打开 | +| Agent | 官方 `deepseek-ai/deepseek-harness`,版本 `0.2.2-alpha.2`,源码 `f3803c96974ad4a94195fb97e5c637e9c2a3afe3` | +| 配置 | 上游推荐用于基准测试的 `sdk-minimal`,通过公开 Python SDK 驱动 | +| 行为 | 原生系统提示词、持久 bash、图片读取、工作目录工具、上下文压缩、请求重试及生成默认值;每题使用全新目录和会话 | +| 隔离 | 不载入个人指令、技能、扩展、设置或历史会话;任务工具在题目容器内执行 | +| 模型 | `deepseek/deepseek-flash`,`high` 推理,官方 Messages 端点 `https://api.deepseek.com/anthropic` | +| 生成设置 | 不设置 temperature 和 SDK `max_tokens`;上游输出上限为 256,000 | +| Adapter | 冻结源码 `5b503ab6d04af2f6a719ec0dbe5418bfad4bf128`,位于 `jobs/deepseek-harness-runtime-source` | +| 限制 | 保留题目原生 agent / 验证器时限和资源;安装时限为 1,800 秒 | +| 队列 | 与 Pi 顺序一致;并发为二,每对题结束后才进入下一对;不重试整题 | +| 验证 | 保留原始脚本与每次验证时限;仅验证器超时可在同一 agent 输出上重试,总共最多三次 | + +10 月 30 日和 31 日检查的官方 +[agent 接入指南](https://api-docs.deepseek.com/quick_start/agent_integrations/) +将 `deepseek-flash` 映射到 DeepSeek-V4.1-Flash-1017; +[别名记录](../../configs/deepseek-harness-dev34-20261030/model-alias-check.json) +保存了后一次检查。它仍是可变别名,不能证明不同日期使用完全相同的后端 +检查点。Pi 使用其原生 OpenAI 兼容端点和 384,000 token 输出上限。两次实验 +均未施加 Nano 的 100 次回复限制或收尾提醒。端点、预算、日期和生命周期 +差异意味着本实验比较的是完整配置。 + +用户原始的上游检出保持干净且未修改。构建在隔离副本中进行,使用上游冻结 +的 pnpm 锁文件(Node 26.7.0、pnpm 11.28.5)。八项构建关卡和官方无密钥 +SDK 冒烟测试全部通过。离线分发包含 SDK、原生运行时和五个依赖 wheel。 +源码内容、分发内容及文件权限均已固定。 + +```text +plan SHA256: fb59d5618c44e691babb80b2ea8323eb45edf9f45383e61344d747b870bc6d6e +freeze SHA256: cc421c1429450ac41d355d0df8333f4ab10ca4da0bf47f24bc3892d0724d9c24 +results SHA256: e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396 +selection SHA256: 4572b1cc432a16b9c5d1dd554c8d2189ae80414f386a90bf03829bc4d0c3da85 +source archive SHA256: 0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4 +distribution SHA256: d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6 +runtime SHA256: 28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2 +lifecycle diagnostic SHA256: 4257a96a3415a7d389fb7b3fb10f819341714ba833b0911eb2a6468171e70c47 +``` + +## 安装、执行边界与错误记账 + +**全部 34 道仅安装预检均在首个模型作业开始前通过**,包括 SDK 初始化和 +Linux 进程监督探针。QEMU 的 `qemu-alpine-ssh-v1` 与模型恢复题的 +`pytorch-model-recovery-v1` 依赖配置通过;其余 **32 道明确标记为 +`not_configured`**。 +[模型执行前的元数据修正](../../configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json) +更正了沿用的历史配置数量描述,没有改变任何执行设置。 +[预检证据](../../configs/deepseek-harness-dev34-20261030/preflight-evidence.json) +不保证后续依赖网络的安装仍能成功。 + +首次 WAL 安装冒烟失败,因为 Python `dsh` 启动器即使查询版本也需要显式 +`DSH_HOME`。Adapter 在注册前为版本检查提供了隔离目录,修正后的冒烟通过。 +两次记录均保存在[安装证据](../../configs/deepseek-harness-dev34-20261030/setup-evidence.json), +都不属于计分模型尝试。验证包括该集成修正之前的 280 项 Harbor 测试,以及 +修正之后的 11 项 DeepSeek 专项测试。最终完整 Harbor 测试再次全部通过,共 +280 项。首次调用通过 273 项,另七项本地 HTTP fixture 被沙箱 socket 权限 +阻止;允许本地 socket 后完整重跑,在 17.56 秒内全部通过。 + +**全部 32 次模型执行**都在官方验证前确认了完成或取消边界。没有记录到 +agent 结束之后,或验证器开始时及之后的 SDK 事件。三次 agent 超时分别是 +采样器、模型提取和流水线并行,均使用原生 900 秒时限;取消过程分别停止了 +**10、4、8 个进程**。采样器和模型提取具有有效的官方失败结果;流水线并行 +另有验证器依赖错误,因此不计分。 + +原始奖励保留为 **22 个一、10 个零、两个 null**。未修改的标准汇总包含 +**28 个 `scored`、四个 `execution_error`、两个 `setup_failed`**。 +分析审计将官方断言失败与安装未完成、终止性的 API 错误及验证器错误分开; +不改写原生异常或奖励。 + +| 未计分题目 | 类别与证据 | +| --- | --- | +| pytorch-model-cli | 安装:获取原始 uv 0.9.11 安装器时,在 `releases.astral.sh:443` 发生 curl 35 / `SSL_ERROR_SYSCALL`;原生 `NetworkConnectionError`,未执行模型 | +| pytorch-model-recovery | 安装:在 `verifier_dependencies` 阶段达到 1,800 秒,原生 `AgentSetupTimeoutError`;SDK / 监督检查通过,但本次正式依赖检查未完成;未执行模型 | +| multi-source-data-merger | 验证器:下载或解压 `pyarrow==22.0.0` 时发生 `UV_HTTP_TIMEOUT=30` 超时;官方 pytest 未开始 | +| query-optimize | 执行:SDK `finish_reason=error`,同一提供方请求尝试六次(含五次原生请求重试)后因 `TRANSPORT` 错误终止;原生 `NonZeroAgentExitCodeError`;后续官方测试发现 SQL 输出缺失 | +| torch-pipeline-parallelism | 验证器:确认 agent 超时取消后,下载 `torch==2.7.0` 所需的 `nvidia-cusolver-cu12==11.7.1.2` 超时;官方 pytest 未开始 | +| torch-tensor-parallelism | 验证器:首次达到 900 秒;第二次下载 `torch==2.7.0` 时发生 `UV_HTTP_TIMEOUT=30` 超时;官方 pytest 未开始 | + +共有**三次额外验证尝试**:采样器两次达到 900 秒超时,第三次在 26.3 秒内 +完成;张量并行一次达到 900 秒超时,第二次因依赖错误结束。重试使用同一 +输出和原始脚本。依赖失败未触发重试。没有重跑任何完整题目、模型输出或 +安装失败。Agent 内的原生请求重试与这些验证器重试、整题重试分别记账。 + +## 逐题结果 + +`—` 表示无分析评分或未执行,依列含义区分。Pi 列采用审计后评分,而非原始 +奖励。`†` 表示服务生命周期限制;`‡` 表示 Pi 曾访问参考材料的流水线通过结果。 + +| 题目 | Pi 审计评分 | Harness 原始奖励 | Harness 审计结果 | 官方测试通过 / 失败 | Agent 秒数 | +| --- | ---: | ---: | --- | ---: | ---: | +| adaptive-rejection-sampler | — | 0 | 超时;确认停止,失败 | 4 / 5 | 899.4 | +| build-cython-ext | 1 | 1 | 通过 | 11 / 0 | 593.1 | +| cancel-async-tasks | 0 | 0 | 失败 | 5 / 1 | 26.3 | +| cobol-modernization | 1 | 1 | 通过 | 3 / 0 | 172.4 | +| db-wal-recovery | — | 1 | 通过 | 7 / 0 | 34.1 | +| fix-code-vulnerability | 1 | 1 | 通过 | 6 / 0 | 42.1 | +| git-multibranch | 1 | 1 | 通过 | 1 / 0 | 83.5 | +| kv-store-grpc † | 1 | 0 | 失败 | 5 / 2 | 45.8 | +| log-summary-date-ranges | 0 | 0 | 失败 | 1 / 1 | 21.0 | +| merge-diff-arc-agi-task | 1 | 1 | 通过 | 5 / 0 | 110.5 | +| model-extraction-relu-logits | 0 | 0 | 超时;确认停止,失败 | 0 / 1 | 899.4 | +| multi-source-data-merger | 1 | 0 | 未计分:验证器错误 | — | 27.8 | +| nginx-request-logging | 1 | 1 | 通过 | 8 / 0 | 138.0 | +| openssl-selfsigned-cert | 1 | 1 | 通过 | 6 / 0 | 36.2 | +| pypi-server † | 1 | 0 | 失败 | 0 / 1 | 40.2 | +| pytorch-model-cli | 0 | — | 未计分:安装错误 | — | — | +| pytorch-model-recovery | 1 | — | 未计分:安装错误 | — | — | +| qemu-alpine-ssh | 1 | 1 | 通过 | 1 / 0 | 270.2 | +| query-optimize | 0 | 0 | 未计分:API 错误 | 2 / 4 | 231.7 | +| regex-log | 1 | 1 | 通过 | 1 / 0 | 90.3 | +| torch-pipeline-parallelism ‡ | 1 | 0 | 未计分:验证器错误 | — | 899.3 | +| torch-tensor-parallelism | 1 | 0 | 未计分:验证器错误 | — | 131.3 | +| write-compressor | 1 | 1 | 通过 | 3 / 0 | 143.6 | +| constraints-scheduling | 1 | 1 | 通过 | 3 / 0 | 24.3 | +| custom-memory-heap-crash | 1 | 1 | 通过 | 6 / 0 | 131.7 | +| dna-assembly | 0 | 1 | 通过 | 1 / 0 | 944.5 | +| llm-inference-batching-scheduler | 1 | 1 | 通过 | 6 / 0 | 503.8 | +| path-tracing | 1 | 1 | 通过 | 5 / 0 | 229.8 | +| schemelike-metacircular-eval | 1 | 1 | 通过 | 1 / 0 | 902.8 | +| caffe-cifar-10 | 1 | 1 | 通过 | 6 / 0 | 2200.5 | +| circuit-fibsqrt | 1 | 1 | 通过 | 3 / 0 | 461.1 | +| mteb-leaderboard | 1 | 1 | 通过 | 2 / 0 | 1172.2 | +| regex-chess | 1 | 1 | 通过 | 4 / 0 | 1907.7 | +| build-pov-ray | 1 | 1 | 通过 | 3 / 0 | 891.9 | + +测试数量来自最终官方 pytest 汇总,保留参数化用例;不累加 agent 自测或 +嵌套的上游测试。**28 次有效评估**共执行 **118 个用例:107 通过、11 失败、 +零错误或跳过**。因 API 中断而未计分的查询优化题另有**两个通过、四个失败** +用例,其余五道未计分题未进入官方 pytest。 + +| 计分失败题目 | 官方断言或缺失产物 | +| --- | --- | +| adaptive-rejection-sampler | 确认超时取消后,缺少 `Rscript` 和生成的采样文件;缺少 R 时安装 R 是题目明确要求,因此属于任务失败,而不是验证器依赖排除项 | +| cancel-async-tasks | 多余任务的清理消息为零,预期为二 | +| kv-store-grpc † | 5328 端口未监听;连接被拒绝 | +| log-summary-date-ranges | 当天 ERROR 数为 74,预期 370 | +| model-extraction-relu-logits | 确认超时取消后缺少 `/app/stolen_A1.npy` | +| pypi-server † | 从 `localhost:8080` 安装 `vectorops==0.1.0` 失败,本地索引不可达 | + +这些是官方验证发现,并非穷尽根因的证明。 + +## SDK 服务生命周期诊断 + +KV 的 agent 曾成功调用 `SetVal` / `GetVal` 并读取值 123。PyPI 的 agent +启动了 `nohup` 服务、获取了索引,并在独立虚拟环境中安装和执行了该包。 +SDK 上下文关闭后,两个服务均无法被官方验证器访问。 + +固定版本的上游 SDK 在正常关闭时会释放托管终端及其后代进程。共享的 Linux +监督器允许正常完成后保留解题服务,但无法保留已被 SDK 停止的服务。 +冻结后的[诊断](../../configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json) +使用上游无密钥 mock fixture 和同一固定运行时,只有**两个本地 mock 请求, +零次真实模型调用**。通过 `nohup` 启动的 Python HTTP 服务在 **SDK 关闭前 +可达,关闭后不可达**。 +[可复用探针](../../scripts/probe_deepseek_lifecycle.py) 记录了自身、fixture 和 +运行时的哈希: + +```bash +jobs/dsh-harbor-venv/bin/python benchmarks/harbor/scripts/probe_deepseek_lifecycle.py \ + --upstream-source jobs/deepseek-harness-build-f3803c9/source \ + --runtime jobs/deepseek-harness-build-f3803c9/source/dist-exe/deepseek-harness-sdk-runtime-linux-x64 +``` + +该诊断发生在全部基准输出冻结之后。它确认了本次 SDK 集成的生命周期限制, +不能证明换一种生命周期后两道题会通过所有官方断言。两个零仍作为本配置的 +计分失败保留。没有在注册运行中修复生命周期、重启服务或重新评分。 + +## 与 Pi 的对照解释 + +Pi 审计记录为 **26 通过、六失败、两个历史超时不计分**。其采样器边界无效, +WAL 边界未经确认;流水线并行通过结果曾访问公开参考测试和解答。这些限制 +继续保留在 [Pi 报告](pi-deepseek-flash-dev34-20261024.md) 中。 + +共同有效的 26 道题中,**Harness 通过 21 道 / Pi 通过 22 道**。唯一新增 +通过是 `dna-assembly`;两道退步是 `kv-store-grpc` 和 `pypi-server`,均有 +生命周期限制。`db-wal-recovery` 本次通过,但 Pi 未计分,因此不算成对提升。 +采样器本次有有效官方失败,但同样没有有效 Pi 对照。六道本次未计分题也 +不能用于推断提升或退步。共同子集由能否有效评估决定,不是注册 34 道题池的 +无偏替代。这里不提供统计控制下的排名、完整成对通过率,也不声称观察到的 +变化只由 harness 导致。 + +## 材料访问、用量、耗时与复现 + +冻结输出后,针对基准仓库、隐藏路径和环境访问的定向扫描覆盖了 **675 次 +记录的工具调用**及匹配候选的输出。该扫描未观察到从外部获取 Terminal-Bench +参考材料或访问官方隐藏测试。日历 `PRODID` 和题目自带 C++ canary 中的 +Terminal-Bench 字符串属于正常匹配。MTEB 下载了公开的 +`embeddings-benchmark/results` 应用数据,版本为 +`3125e8302256ac01d4dd68b6f92902b3e6a78987`,与 Terminal-Bench 测试或解答 +不同。该开发题池历史上已被接触;定向扫描不能证明完全不存在材料暴露或 +数据外传。 + +对全部 **2,506 个冻结文件**按字节精确扫描已授权的提供方凭据,发现 +**零个文件**包含该值。结论仅限于该密钥及这些文件,不改变历史 Pi 的凭据 +发现。原始轨迹和凭据值不发布。维护者 10 月 31 日的授权和模型启动情况 +记录在[启动证据](../../configs/deepseek-harness-dev34-20261030/launch-evidence.json)。 + +| 已知用量或耗时 | 值 | +| --- | ---: | +| 输入 token,含缓存输入 | 25,052,194 | +| 缓存输入 token,已包含在上一项 | 24,346,112 | +| 输出 token | 634,263 | +| 费用 | 未知 | +| Agent 执行秒数之和 | 14,306.464573 | +| 队列墙钟秒数,含安装和验证 | 16,984.894313 | +| 首个作业开始(UTC) | 2026-10-31 11:03:13.156046 | +| 最后作业结束(UTC) | 2026-10-31 15:46:18.050359 | + +用量为**已知部分合计**。采样器、模型提取、查询优化、流水线并行、DNA +组装和 LLM 批处理均存在部分用量统计。失败的原生重试与被中断请求可能 +缺失用量,即使之后成功完成也一样。不要再次把缓存 token 加到输入中,也 +不能据此推断账单。网络状况、安装失败、验证器重试及有效题目集合不同, +无法进行受控的速度或费用对比。 + +最终验证重新检查了源码、题目和配置哈希、全部预检先于模型的边界、每个 +双题屏障,以及预检和正式作业中的 **68 次原始镜像检查**。原始软件包来源 +和题目镜像均保留;缓存命中必须匹配文件名、大小和 SHA256。仅对本次新增 +镜像引用进行非强制清理,保留受保护镜像。34 个作业均已结束,**零个题目 +容器仍在运行**。 + +[配置 README](../../configs/deepseek-harness-dev34-20261030/README.md) 包含 +构建和执行命令。每个作业均从冻结源码使用标准 +`uv run --project benchmarks/harbor python -m harbor_adapter run --config ...` +入口。原生证据保留在 `jobs/deepseek-harness-runtime-source/jobs/`,编排和 +镜像事件保留在 `jobs/deepseek-harness-dev34-20261030-orchestration/`。 +发布的 JSON 和哈希清单支持检查来源;从源码复现还需要注册的题目包、镜像、 +已构建分发及提供方访问权限。已注册的现有作业拒绝覆盖或重放。 diff --git a/benchmarks/harbor/results/deepseek-harness-dev34-20261030.json b/benchmarks/harbor/results/deepseek-harness-dev34-20261030.json new file mode 100644 index 0000000..5402054 --- /dev/null +++ b/benchmarks/harbor/results/deepseek-harness-dev34-20261030.json @@ -0,0 +1,5878 @@ +{ + "schema_version": 1, + "experiment": "deepseek-harness-dev34-20261030", + "audited_at": "2026-10-31T16:01:42.398639+00:00", + "frozen_at": "2026-10-31T15:47:11.582274+00:00", + "freeze_manifest": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json", + "freeze_sha256": "cc421c1429450ac41d355d0df8333f4ab10ca4da0bf47f24bc3892d0724d9c24", + "registration": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/experiment.json", + "registration_sha256": "fb59d5618c44e691babb80b2ea8323eb45edf9f45383e61344d747b870bc6d6e", + "selection_sha256": "4572b1cc432a16b9c5d1dd554c8d2189ae80414f386a90bf03829bc4d0c3da85", + "scope": "Exact previously exposed Pi development pool of 34 tasks, unchanged revisions, images, resources, official tests and native limits. The other 55 tasks remained unopened. 34 formal jobs, 32 model executions, no replacement tasks or whole-trial replay.", + "model_policy": { + "model": "deepseek/deepseek-flash", + "provider": "deepseek-official", + "api": "anthropic-messages", + "base_url": "https://api.deepseek.com/anthropic", + "reasoning_effort": "high", + "profile": "sdk-minimal", + "temperature": "unset", + "sdk_max_tokens": "unset", + "upstream_default_max_tokens": 256000, + "alias": "Official documentation checked 2026-10-30 maps the mutable deepseek-flash alias to DeepSeek-V4.1-Flash-1017." + }, + "budget_policy": "Original task-native timeouts/resources; 1800-second setup. Upstream sdk-minimal prompts, tools, compaction, retries, and generation defaults. No injected turn limit, internal time budget, or finalization reminders. Pi used OpenAI-compatible transport with 384000-token model metadata; this is not a matched-generation-budget causal harness comparison.", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "sources": { + "adapter": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "upstream": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "upstream_origin": "https://github.com/deepseek-ai/deepseek-harness.git", + "upstream_version": "0.2.2-alpha.2", + "source_archive_sha256": "0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4", + "runtime_sha256": "28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2", + "profile": "sdk-minimal" + }, + "summary": { + "registered_jobs": 34, + "model_executions": 32, + "passed": 22, + "failed": 6, + "unscored": 6, + "analytical_classifications": { + "failed": 6, + "passed": 22, + "unscored_verifier_error": 3, + "unscored_setup_error": 2, + "unscored_execution_error": 1 + }, + "native_rewards": { + "one": 22, + "zero": 10, + "null": 2 + }, + "standard_summary_statuses": { + "execution_error": 4, + "scored": 28, + "setup_failed": 2 + }, + "agent_timeouts": 3, + "whole_trial_retries": 0, + "extra_verifier_attempts": 3, + "service_lifecycle_caveat_failures": 2, + "accepted_official_cases": { + "total": 118, + "passed": 107, + "failed": 11, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "unscored_query_official_cases": { + "passed": 2, + "failed": 4 + } + }, + "accounting_policy": "Native rewards, exceptions and standard summaries are unchanged. Confirmed timeout outputs with completed official tests remain scoreable; setup failures, terminal provider transport failure and verifier dependency failures remain analytically null. Service-lifecycle failures are retained as failures of this measured SDK configuration. This is not a complete 34-task pass rate or a causal agent-quality comparison.", + "usage": { + "n_input_tokens_known": 25052194, + "n_cache_tokens_known": 24346112, + "n_output_tokens_known": 634263, + "input_includes_cache": true, + "cost_usd": null, + "usage_is_partial": true, + "partial_usage_tasks": [ + "terminal-bench/adaptive-rejection-sampler", + "terminal-bench/model-extraction-relu-logits", + "terminal-bench/query-optimize", + "terminal-bench/torch-pipeline-parallelism", + "terminal-bench/dna-assembly", + "terminal-bench/llm-inference-batching-scheduler" + ], + "policy": "Known SDK usage subtotals only. Interrupted requests and failed native retries can lack usage. Cache tokens are already included in input; do not add them again. No provider billing or price-derived cost is asserted." + }, + "timing": { + "first_job_started_at": "2026-10-31T11:03:13.156046+00:00", + "last_job_finished_at": "2026-10-31T15:46:18.050359+00:00", + "queue_wall_seconds": 16984.894313, + "summed_agent_seconds": 14306.464573, + "interpretation": "Queue wall time includes setup, verifier retries, and pair barriers. Dates, API failures, package downloads and different accepted task sets prevent a controlled latency comparison." + }, + "historical_comparison": { + "baseline": { + "path": "benchmarks/harbor/results/pi-deepseek-flash-dev34-20261024.json", + "sha256": "b85d3bf2b0ea9661940f0e6b08e3bea71b2c41ae34da5c89f9ecd5197c6c8ff5" + }, + "pi_audited": { + "passed": 26, + "failed": 6, + "unscored": 2 + }, + "common_accepted_tasks": 26, + "deepseek_passes_on_common": 21, + "pi_passes_on_common": 22, + "gained_passes": [ + "terminal-bench/dna-assembly" + ], + "lost_passes": [ + "terminal-bench/kv-store-grpc", + "terminal-bench/pypi-server" + ], + "excluded_from_paired_comparison": [ + "terminal-bench/adaptive-rejection-sampler", + "terminal-bench/db-wal-recovery", + "terminal-bench/multi-source-data-merger", + "terminal-bench/pytorch-model-cli", + "terminal-bench/pytorch-model-recovery", + "terminal-bench/query-optimize", + "terminal-bench/torch-pipeline-parallelism", + "terminal-bench/torch-tensor-parallelism" + ], + "caveats": "Both losses have the confirmed SDK service-lifecycle caveat. WAL passes here but was unscored in Pi and is not a paired gain. Pi sampler and WAL have legacy invalid/unverified boundaries; Pi pipeline pass has public reference-test/solution exposure. Different dates, mutable alias, Messages versus OpenAI-compatible endpoints and 256000 versus 384000 output ceilings prevent causal harness attribution." + }, + "validation": { + "final_plan_validation_passed": true, + "source_task_and_configuration_hashes_match": true, + "all_34_preflights_before_first_model_job": true, + "pair_barriers_passed": true, + "registered_jobs_finished": 34, + "model_executions": 32, + "active_task_containers_after_run": 0, + "images_verified": 68, + "first_job_started_at": "2026-10-31T11:03:13.156046+00:00", + "last_job_finished_at": "2026-10-31T15:46:18.050359+00:00", + "frozen_files_rehashed": 2506, + "process_boundaries_confirmed_before_verification": 32, + "post_agent_end_events": 0, + "events_at_or_after_verifier_start": 0, + "official_task_files_match_registered_hashes": true, + "final_harbor_tests_passed": 280, + "final_harbor_test_command": "PYTHONPATH=benchmarks/harbor/src jobs/dsh-harbor-venv/bin/python -m pytest benchmarks/harbor/tests -q", + "final_harbor_test_note": "273 passed and 7 local HTTP fixtures were blocked by sandbox socket permissions; all 280 passed with local sockets permitted.", + "bilingual_reports_matching_task_rows": 34, + "bilingual_report_code_blocks_hashes_and_numeric_cells_match": true, + "publication_files_exact_credential_matches": 0, + "keyless_lifecycle_probe_passed": true + }, + "preflights": [ + { + "task": "terminal-bench/adaptive-rejection-sampler", + "task_ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de", + "job": "dsh-dev34-install-pair-01-20261030", + "trial": "adaptive-rejection-sampler__BuqYqRV", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-01-20261030.json", + "sha256": "35b97de60af32fd25dcb0223dfc5cade5f491cc0361a2424e7973cde21826860" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/manifest.json", + "sha256": "a47999c0d83ce04d3b6ace156f7e73b9763cd4e495e9c51708547b99285a07ca" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/summary.json", + "sha256": "61bae69a5c87321a32a782a96fcd43e627d9c45049a7cf9cb24648e5d1143a31" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/result.json", + "sha256": "9543b465485a93422a68d7741ee610603545d2e9b41cec779106bdef39c4d225" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/adaptive-rejection-sampler__BuqYqRV/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/build-cython-ext", + "task_ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12", + "job": "dsh-dev34-install-pair-01-20261030", + "trial": "build-cython-ext__4W44WcX", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-01-20261030.json", + "sha256": "35b97de60af32fd25dcb0223dfc5cade5f491cc0361a2424e7973cde21826860" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/manifest.json", + "sha256": "a47999c0d83ce04d3b6ace156f7e73b9763cd4e495e9c51708547b99285a07ca" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030-input/summary.json", + "sha256": "61bae69a5c87321a32a782a96fcd43e627d9c45049a7cf9cb24648e5d1143a31" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/result.json", + "sha256": "cc060ca2421d5d182bc554c9f5f9f003521ec0f4a25b9fc3fbce0bc6e0edc8f7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-01-20261030/build-cython-ext__4W44WcX/agent/bootstrap.json", + "sha256": "0c82eb159d150060290eebcfbe93607927cd4537bc16959451b2dcc648d61767" + } + ] + }, + { + "task": "terminal-bench/cancel-async-tasks", + "task_ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1", + "job": "dsh-dev34-install-pair-02-20261030", + "trial": "cancel-async-tasks__cagFx9t", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-02-20261030.json", + "sha256": "b5a81d0535ecaaac098b9b5b4839c5d5946950d70de5aba646fd27bd700320d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/manifest.json", + "sha256": "0a7a13cc673070c646be8d06fd8fbd2013c06f32a1fff6131d03d1c77755f95c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/summary.json", + "sha256": "1d57bea97ec7cd5d3cc348e065195ac5579e7d1a3786a177b8b3c5a0e2389a42" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/result.json", + "sha256": "880f693138aae3abba7bfdb0a178aaadbc8c9629e329656d4ccda8ee84a93d1e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cancel-async-tasks__cagFx9t/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/cobol-modernization", + "task_ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88", + "job": "dsh-dev34-install-pair-02-20261030", + "trial": "cobol-modernization__Uorq3pW", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-02-20261030.json", + "sha256": "b5a81d0535ecaaac098b9b5b4839c5d5946950d70de5aba646fd27bd700320d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/manifest.json", + "sha256": "0a7a13cc673070c646be8d06fd8fbd2013c06f32a1fff6131d03d1c77755f95c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030-input/summary.json", + "sha256": "1d57bea97ec7cd5d3cc348e065195ac5579e7d1a3786a177b8b3c5a0e2389a42" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/result.json", + "sha256": "92b26a3a93ff64857ecd4b6d65b6853308aefb3eb3f190614967d1a790f3a567" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-02-20261030/cobol-modernization__Uorq3pW/agent/bootstrap.json", + "sha256": "d75e36e5346e74a3f1b464b24b6e3bbf159103f626e474d7833ffed5bb202221" + } + ] + }, + { + "task": "terminal-bench/db-wal-recovery", + "task_ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4", + "job": "dsh-dev34-install-pair-03-20261030", + "trial": "db-wal-recovery__SsWt56x", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-03-20261030.json", + "sha256": "7997484bc8cf759eac5ad91dcbc9aee1a310d1ab9de5d962aa2dab6f4141e2ad" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/manifest.json", + "sha256": "2f35cbc799352d952caabace8f9ed3eb9f9d6bea62f390088e9c92698362d61e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/summary.json", + "sha256": "90943816f7581a3676aa4a7c4f6052ec055de61c1de145e8c58c17af03fc400f" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/result.json", + "sha256": "625c9a6e005e000984e7fc8111f481fd86ee259893d3c8f5cffe18b4b6291160" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/db-wal-recovery__SsWt56x/agent/bootstrap.json", + "sha256": "317eb4d1a905a9446f48890f4be51b9acf3c0e230c0770b360285bae70d579b1" + } + ] + }, + { + "task": "terminal-bench/fix-code-vulnerability", + "task_ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71", + "job": "dsh-dev34-install-pair-03-20261030", + "trial": "fix-code-vulnerability__7tYLPYo", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-03-20261030.json", + "sha256": "7997484bc8cf759eac5ad91dcbc9aee1a310d1ab9de5d962aa2dab6f4141e2ad" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/manifest.json", + "sha256": "2f35cbc799352d952caabace8f9ed3eb9f9d6bea62f390088e9c92698362d61e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030-input/summary.json", + "sha256": "90943816f7581a3676aa4a7c4f6052ec055de61c1de145e8c58c17af03fc400f" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/result.json", + "sha256": "26f8e8a79707a4f334422558031a0e2faafef53aad085ca04a688f8a2623a926" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-03-20261030/fix-code-vulnerability__7tYLPYo/agent/bootstrap.json", + "sha256": "a243941e9357e36a407ff4cfd4cf32498b67756cf083283184e3ac20269ed125" + } + ] + }, + { + "task": "terminal-bench/git-multibranch", + "task_ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa", + "job": "dsh-dev34-install-pair-04-20261030", + "trial": "git-multibranch__uksBvS7", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-04-20261030.json", + "sha256": "08cc0bb602fdc5741e4432833ada51081e2d311d3030c63b700961c658545130" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/manifest.json", + "sha256": "e46c64ccb95f0ed5b47e11db1c4404df71ab21a2096ddfc3df560a3e3c8a36d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/summary.json", + "sha256": "1262258e1c174e5c3e0cae701841e515c4e61714297ad141f6ba5d7ce637ce4a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/result.json", + "sha256": "aa8f47a7aa4d3e558ef34fb2ee22f9ec7cdcf7c44f4eb0a54c64e5d204384ac5" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/git-multibranch__uksBvS7/agent/bootstrap.json", + "sha256": "139a6fd0776dee86f5a0306ebdadb3f1a4e79938a27aa2a5edcdd96dfa9be907" + } + ] + }, + { + "task": "terminal-bench/kv-store-grpc", + "task_ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307", + "job": "dsh-dev34-install-pair-04-20261030", + "trial": "kv-store-grpc__V6nNm5R", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-04-20261030.json", + "sha256": "08cc0bb602fdc5741e4432833ada51081e2d311d3030c63b700961c658545130" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/manifest.json", + "sha256": "e46c64ccb95f0ed5b47e11db1c4404df71ab21a2096ddfc3df560a3e3c8a36d1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030-input/summary.json", + "sha256": "1262258e1c174e5c3e0cae701841e515c4e61714297ad141f6ba5d7ce637ce4a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/result.json", + "sha256": "ed6e15f3638c215e50b21a7dce5a86eebcbbb07ad5778fa02eace1d89371baad" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-04-20261030/kv-store-grpc__V6nNm5R/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/log-summary-date-ranges", + "task_ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6", + "job": "dsh-dev34-install-pair-05-20261030", + "trial": "log-summary-date-ranges__6dcMfYE", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-05-20261030.json", + "sha256": "b75fd05b46d16329753c582920cd1b08b75695f6a4728f293ecbc6bb2dcd20ed" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/manifest.json", + "sha256": "09adca648d3cd50c864e95c0b3d36326b7db09b5227435580427cccd893102bd" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/summary.json", + "sha256": "0196f9f1298444e0a42ff5070ccd7f89717ba14cb8bc5edb34ec404e066de7b4" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/result.json", + "sha256": "2c5d6232aeec9ee5b4c040a0a9ef0e5e76f360bf880579cba4bae9d10b69ad5d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/log-summary-date-ranges__6dcMfYE/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/merge-diff-arc-agi-task", + "task_ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d", + "job": "dsh-dev34-install-pair-05-20261030", + "trial": "merge-diff-arc-agi-task__RGHcv6Z", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-05-20261030.json", + "sha256": "b75fd05b46d16329753c582920cd1b08b75695f6a4728f293ecbc6bb2dcd20ed" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/manifest.json", + "sha256": "09adca648d3cd50c864e95c0b3d36326b7db09b5227435580427cccd893102bd" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030-input/summary.json", + "sha256": "0196f9f1298444e0a42ff5070ccd7f89717ba14cb8bc5edb34ec404e066de7b4" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/result.json", + "sha256": "2cdf1b3a5485cdbc6db2f390c5186dec2939d374c476543f04edad59a0c66a14" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-05-20261030/merge-diff-arc-agi-task__RGHcv6Z/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/model-extraction-relu-logits", + "task_ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17", + "job": "dsh-dev34-install-pair-06-20261030", + "trial": "model-extraction-relu-logits__Ky6SN6a", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-06-20261030.json", + "sha256": "f7c8461d95344962910447b29e0b4b942794d2aeb5d854e245100ba2f7b5a45b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/manifest.json", + "sha256": "c0f6916a703f0c9110e958bf0f797a2b4a261f65c6be3fc137d3a54daf01db73" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/summary.json", + "sha256": "b13d0543f0cb1c0f367e6f74317c2fe759e399297e2061665719c64847dfdb22" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/result.json", + "sha256": "1cf09cef6d5f195c6f64e1dddf49fe52fe61583e8a6ffaf75bb9e6454ae2c99e" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/model-extraction-relu-logits__Ky6SN6a/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/multi-source-data-merger", + "task_ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "job": "dsh-dev34-install-pair-06-20261030", + "trial": "multi-source-data-merger__pxBy43s", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-06-20261030.json", + "sha256": "f7c8461d95344962910447b29e0b4b942794d2aeb5d854e245100ba2f7b5a45b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/manifest.json", + "sha256": "c0f6916a703f0c9110e958bf0f797a2b4a261f65c6be3fc137d3a54daf01db73" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030-input/summary.json", + "sha256": "b13d0543f0cb1c0f367e6f74317c2fe759e399297e2061665719c64847dfdb22" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/result.json", + "sha256": "cad316ea002c327d02d2eebdcb379f5cc2b631396cefc82cb34d306f4e92c16c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-06-20261030/multi-source-data-merger__pxBy43s/agent/bootstrap.json", + "sha256": "38a018af813827b21bff5685e056300a44dc7899ae803e02ae564568808be3fd" + } + ] + }, + { + "task": "terminal-bench/nginx-request-logging", + "task_ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097", + "job": "dsh-dev34-install-pair-07-20261030", + "trial": "nginx-request-logging__ra8o6iV", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-07-20261030.json", + "sha256": "8dc0becd58a3de76db3fbb638e01e6a0296fd3680cc37c50cec9288f94d57d41" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/manifest.json", + "sha256": "59297ca64e29863872cfa97ab882a77328d38ad7f5584a6890489c8825f81800" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/summary.json", + "sha256": "37b07dd932adf6c214a49be3fcc88a67ee069f1d178d036b372829d6ffe85a48" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/result.json", + "sha256": "5f06c592c06135e40b91172af316333c4edca880235a1f386784ea97acc24d59" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/nginx-request-logging__ra8o6iV/agent/bootstrap.json", + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e" + } + ] + }, + { + "task": "terminal-bench/openssl-selfsigned-cert", + "task_ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302", + "job": "dsh-dev34-install-pair-07-20261030", + "trial": "openssl-selfsigned-cert__6stQ3ja", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-07-20261030.json", + "sha256": "8dc0becd58a3de76db3fbb638e01e6a0296fd3680cc37c50cec9288f94d57d41" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/manifest.json", + "sha256": "59297ca64e29863872cfa97ab882a77328d38ad7f5584a6890489c8825f81800" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030-input/summary.json", + "sha256": "37b07dd932adf6c214a49be3fcc88a67ee069f1d178d036b372829d6ffe85a48" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/result.json", + "sha256": "6b07e89c5e841b478bfc4241ecf3808dd391f9cff3e3e95dff16d50a6d28823b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-07-20261030/openssl-selfsigned-cert__6stQ3ja/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/pypi-server", + "task_ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1", + "job": "dsh-dev34-install-pair-08-20261030", + "trial": "pypi-server__4jBWBVR", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-08-20261030.json", + "sha256": "9d4230aae5324a094c5d2ec43236b6cb651cd777805d46821527cdd8c607d4eb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/manifest.json", + "sha256": "97bdcff4c34faf3d787c3860d68f8b958812d6ae51f20ed00012b62abb4da721" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/summary.json", + "sha256": "53288df2ea90779162c20bd6ebdfc7b38a8d38f90acf3794a9e47f2ad711e2a7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/result.json", + "sha256": "73709f5a967b2caf1082b5a45144fa483bfdb0643f599e792db0ad5da11f3a9c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pypi-server__4jBWBVR/agent/bootstrap.json", + "sha256": "1ca2baf6b033a6a37c7bfa6b0baa648f7b5784697774fa5af801e7ac35f2f27e" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-cli", + "task_ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "job": "dsh-dev34-install-pair-08-20261030", + "trial": "pytorch-model-cli__Lnfimqu", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-08-20261030.json", + "sha256": "9d4230aae5324a094c5d2ec43236b6cb651cd777805d46821527cdd8c607d4eb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/manifest.json", + "sha256": "97bdcff4c34faf3d787c3860d68f8b958812d6ae51f20ed00012b62abb4da721" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030-input/summary.json", + "sha256": "53288df2ea90779162c20bd6ebdfc7b38a8d38f90acf3794a9e47f2ad711e2a7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/result.json", + "sha256": "a50ce1fd10a899155353e312aa5ef356fb05194cfba6f73e899c9e9da4fec620" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-08-20261030/pytorch-model-cli__Lnfimqu/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "task_ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "job": "dsh-dev34-install-pair-09-20261030", + "trial": "pytorch-model-recovery__FCnSvrf", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "passed" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-09-20261030.json", + "sha256": "cee92389b124bfbca5c1d05a1c3cd1467b2986ef9a6bef7248537dcc3ba81f93" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/manifest.json", + "sha256": "00a879f86f550d9aea4530e2fd058f04890540a26c0b2a15f44b2d5f50b01803" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/summary.json", + "sha256": "4f447719c1b8d78ab9aede92dddd9145abfbfb292efc75b40e7cffac7184ec45" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/result.json", + "sha256": "6a7bf4ba88b91f90a6cd845403006de53b7ac3ad2df1db84d8f09ca0125730ef" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/pytorch-model-recovery__FCnSvrf/agent/bootstrap.json", + "sha256": "b97b124b61cfc3b413b29b69460372d0b964474f0b96133c4a5571d9df639c45" + } + ] + }, + { + "task": "terminal-bench/qemu-alpine-ssh", + "task_ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28", + "job": "dsh-dev34-install-pair-09-20261030", + "trial": "qemu-alpine-ssh__eb5Y6dk", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "qemu-alpine-ssh-v1", + "status": "passed" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-09-20261030.json", + "sha256": "cee92389b124bfbca5c1d05a1c3cd1467b2986ef9a6bef7248537dcc3ba81f93" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/manifest.json", + "sha256": "00a879f86f550d9aea4530e2fd058f04890540a26c0b2a15f44b2d5f50b01803" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030-input/summary.json", + "sha256": "4f447719c1b8d78ab9aede92dddd9145abfbfb292efc75b40e7cffac7184ec45" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/result.json", + "sha256": "0e3ccd245745a31c01eadf8d10e4ae5a061d083fc07e828211a99560cf9196c3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-09-20261030/qemu-alpine-ssh__eb5Y6dk/agent/bootstrap.json", + "sha256": "1e672441c98d4e4e39bbb460a3c2c7e1d185954628be757adf5af4f0eef9a767" + } + ] + }, + { + "task": "terminal-bench/query-optimize", + "task_ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "job": "dsh-dev34-install-pair-10-20261030", + "trial": "query-optimize__AbiSxXy", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-10-20261030.json", + "sha256": "b70a073249ebb1d2a688c3a6f5f068519f67beef0423e5179f966bc93e866138" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/manifest.json", + "sha256": "502459c6482d86928087898c676b5390c2229d0009c952fb02da783352bcf51a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/summary.json", + "sha256": "d6654fb376206566eb2d98c82cbb9f0a3ecbe0047456c5e74878530ca465e258" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/result.json", + "sha256": "6fa73803640f1d1bc557ef8bac23af91149d7f55c10bb5c7112e45239115812c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/query-optimize__AbiSxXy/agent/bootstrap.json", + "sha256": "89006797609af114ee0903b557c5ff434b3ef6071d8346100829e653fceb38ca" + } + ] + }, + { + "task": "terminal-bench/regex-log", + "task_ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a", + "job": "dsh-dev34-install-pair-10-20261030", + "trial": "regex-log__VD4fbS3", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-10-20261030.json", + "sha256": "b70a073249ebb1d2a688c3a6f5f068519f67beef0423e5179f966bc93e866138" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/manifest.json", + "sha256": "502459c6482d86928087898c676b5390c2229d0009c952fb02da783352bcf51a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030-input/summary.json", + "sha256": "d6654fb376206566eb2d98c82cbb9f0a3ecbe0047456c5e74878530ca465e258" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/result.json", + "sha256": "091663f68b91509c2337d427b83e75b3b4a7df9bfcad6e94afd4ec565831af3f" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-10-20261030/regex-log__VD4fbS3/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "task_ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "job": "dsh-dev34-install-pair-11-20261030", + "trial": "torch-pipeline-parallelism__AmhPcjR", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-11-20261030.json", + "sha256": "0b35669d6acce8ac74eb247f50d00c36a944b274020be486bfd49821eb7abcc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/manifest.json", + "sha256": "8b148e8e8bc4495d1806a39c7e4e37d1690172ff925179890fd0918f4583365d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/summary.json", + "sha256": "2014331be4e6536020bbba6ae8705958c391f21cf40b4cbf45d042eebad3a727" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/result.json", + "sha256": "bc184f6f88666ccdfe68e25811e45b5ebcacc321b247307bb88bff5408efd968" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-pipeline-parallelism__AmhPcjR/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "task_ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "job": "dsh-dev34-install-pair-11-20261030", + "trial": "torch-tensor-parallelism__NVe5csr", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-11-20261030.json", + "sha256": "0b35669d6acce8ac74eb247f50d00c36a944b274020be486bfd49821eb7abcc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/manifest.json", + "sha256": "8b148e8e8bc4495d1806a39c7e4e37d1690172ff925179890fd0918f4583365d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030-input/summary.json", + "sha256": "2014331be4e6536020bbba6ae8705958c391f21cf40b4cbf45d042eebad3a727" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/result.json", + "sha256": "5ee413a851303fa81a99928b7a8b2653bf933f2452672b1b52841cc827edaf7c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-11-20261030/torch-tensor-parallelism__NVe5csr/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/constraints-scheduling", + "task_ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53", + "job": "dsh-dev34-install-pair-12-20261030", + "trial": "constraints-scheduling__4EcXjX2", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-12-20261030.json", + "sha256": "bef506c7122cdd23faccdea2f9631b32c7dbc89fbdf961313f822c9aaee85c7a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/manifest.json", + "sha256": "93bcf89c9a1af9248e825cca85eb2bade9579b60a8a79fb12ff0ce7e8870537d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/summary.json", + "sha256": "6f527aa51e0909c36993a35e0784ed31c01418d97d8355d745e979cda8351577" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/result.json", + "sha256": "6aacea30f7ff30f3913b0b76892263bf23add9253d86e0f76ea0b8fe4b1067fd" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/constraints-scheduling__4EcXjX2/agent/bootstrap.json", + "sha256": "cf2881c5ea9a508fcf319b2a6befc60d39bd25004a7d4f3d2b194579b31fe401" + } + ] + }, + { + "task": "terminal-bench/write-compressor", + "task_ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e", + "job": "dsh-dev34-install-pair-12-20261030", + "trial": "write-compressor__HFAgpjW", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-12-20261030.json", + "sha256": "bef506c7122cdd23faccdea2f9631b32c7dbc89fbdf961313f822c9aaee85c7a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/manifest.json", + "sha256": "93bcf89c9a1af9248e825cca85eb2bade9579b60a8a79fb12ff0ce7e8870537d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030-input/summary.json", + "sha256": "6f527aa51e0909c36993a35e0784ed31c01418d97d8355d745e979cda8351577" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/result.json", + "sha256": "1e5a5194aedd3712d9af22c4a6c25124068a632a8e8846e7c7700820acada7a1" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-12-20261030/write-compressor__HFAgpjW/agent/bootstrap.json", + "sha256": "106fca350e150fcc87b6003fc699eec6e057797103b83a4c169cf5c14a6b293c" + } + ] + }, + { + "task": "terminal-bench/custom-memory-heap-crash", + "task_ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e", + "job": "dsh-dev34-install-pair-13-20261030", + "trial": "custom-memory-heap-crash__DQuhNvN", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-13-20261030.json", + "sha256": "88e9bcb621608f17e1c371707074425eab1e5e60a56c7999eced6cbdf394cb65" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/manifest.json", + "sha256": "eb140223825530b724c6b4a7455a51a44b4a6d9a0c0048a20bab069bf7111446" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/summary.json", + "sha256": "995297252ae987cf60c3d4b780ab768e4b51f1bf0bdb0c2327fddcf92ba9547c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/result.json", + "sha256": "8e1e193dd8d4b1bfeb8cb65a45a488e946deabf848ea8322e28bc630c0465449" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/custom-memory-heap-crash__DQuhNvN/agent/bootstrap.json", + "sha256": "747599891fefaac25e7ab1119fee62bb88a9bf9b3e036a496c37c79d15610b74" + } + ] + }, + { + "task": "terminal-bench/dna-assembly", + "task_ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f", + "job": "dsh-dev34-install-pair-13-20261030", + "trial": "dna-assembly__38WuERG", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-13-20261030.json", + "sha256": "88e9bcb621608f17e1c371707074425eab1e5e60a56c7999eced6cbdf394cb65" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/manifest.json", + "sha256": "eb140223825530b724c6b4a7455a51a44b4a6d9a0c0048a20bab069bf7111446" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030-input/summary.json", + "sha256": "995297252ae987cf60c3d4b780ab768e4b51f1bf0bdb0c2327fddcf92ba9547c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/result.json", + "sha256": "b005dd171d4df9c4d790f4518fad8122370217d3252fa14b9bd6cd4c21508aa7" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-13-20261030/dna-assembly__38WuERG/agent/bootstrap.json", + "sha256": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + } + ] + }, + { + "task": "terminal-bench/llm-inference-batching-scheduler", + "task_ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67", + "job": "dsh-dev34-install-pair-14-20261030", + "trial": "llm-inference-batching-scheduler__kd5nP2N", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-14-20261030.json", + "sha256": "e6b14ee3f5a53b3bd0d4b6c54b92db8fc78046dfe82ed834985bccc8bdb56c64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/manifest.json", + "sha256": "86d588aafeabba43bd931a297bc78b960761b2f9b878bece05301deab3defad3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/summary.json", + "sha256": "2cef49e9735b839280086ea02251978e776fbebbde556d0bb5ecc2dcbebbaa64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/result.json", + "sha256": "d9d86f9778f07453b8949b1eafe18f8349d059f247d623410e90f63c9a504561" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/llm-inference-batching-scheduler__kd5nP2N/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/path-tracing", + "task_ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c", + "job": "dsh-dev34-install-pair-14-20261030", + "trial": "path-tracing__2oz9xrm", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-14-20261030.json", + "sha256": "e6b14ee3f5a53b3bd0d4b6c54b92db8fc78046dfe82ed834985bccc8bdb56c64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/manifest.json", + "sha256": "86d588aafeabba43bd931a297bc78b960761b2f9b878bece05301deab3defad3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030-input/summary.json", + "sha256": "2cef49e9735b839280086ea02251978e776fbebbde556d0bb5ecc2dcbebbaa64" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/result.json", + "sha256": "0878bb05a625a7f36d0f90fd39f02e30e522daa6af9781399da7f34930b0aedb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-14-20261030/path-tracing__2oz9xrm/agent/bootstrap.json", + "sha256": "8a3916db45911667d182143670fe4edfb134d999147e568a8af01013ff22ca93" + } + ] + }, + { + "task": "terminal-bench/caffe-cifar-10", + "task_ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf", + "job": "dsh-dev34-install-pair-15-20261030", + "trial": "caffe-cifar-10__gofkboQ", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-15-20261030.json", + "sha256": "7d3111e619275c66f4adad6cb420171e1d1c5ae80401b38288dc830874f313de" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/manifest.json", + "sha256": "2c81a93d2ba080cb4bcc3c359eb10e95b4074eb5c324f2076fb5db330fab8942" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/summary.json", + "sha256": "be4267f0ea72f63625f79ab7f8c43a598fb2eb358464564bda289a9b8e3db02b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/result.json", + "sha256": "56e49dc1210e839e68a550346dd1011637a9780345c7a5a9cdf84fea3c360e75" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/caffe-cifar-10__gofkboQ/agent/bootstrap.json", + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d" + } + ] + }, + { + "task": "terminal-bench/schemelike-metacircular-eval", + "task_ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6", + "job": "dsh-dev34-install-pair-15-20261030", + "trial": "schemelike-metacircular-eval__sesEXMa", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-15-20261030.json", + "sha256": "7d3111e619275c66f4adad6cb420171e1d1c5ae80401b38288dc830874f313de" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/manifest.json", + "sha256": "2c81a93d2ba080cb4bcc3c359eb10e95b4074eb5c324f2076fb5db330fab8942" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030-input/summary.json", + "sha256": "be4267f0ea72f63625f79ab7f8c43a598fb2eb358464564bda289a9b8e3db02b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/result.json", + "sha256": "a13296c3bbd5cfb2e2e6e7c0e37d360492178221e49a2cd868078abe876d7629" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-15-20261030/schemelike-metacircular-eval__sesEXMa/agent/bootstrap.json", + "sha256": "8c54bd5f6bda5759a824483e4799970db539ed9cb18f1b1962bf978e392bc02a" + } + ] + }, + { + "task": "terminal-bench/circuit-fibsqrt", + "task_ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3", + "job": "dsh-dev34-install-pair-16-20261030", + "trial": "circuit-fibsqrt__6JWTyKR", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-16-20261030.json", + "sha256": "03782151798a8c423bf3dd80c0381749060f14443fa18573990d0a52cece5bc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/manifest.json", + "sha256": "757978d6b034758a75027aac61264ae6eb72243da9338ffafdeae28cfa78051b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/summary.json", + "sha256": "40055254b4c8127df523ee89e946b0426f1f65b2935643c17bb447d5f40f3a5c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/result.json", + "sha256": "2a6b3a5ec5956d1249a5663543ac1149cb4a1fecfaadd0b7c34a6a1721939c60" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/circuit-fibsqrt__6JWTyKR/agent/bootstrap.json", + "sha256": "933708cbc5cd3c97f336c849d1eee8af9f3b824eddd6aa7506dc53f1657d8247" + } + ] + }, + { + "task": "terminal-bench/mteb-leaderboard", + "task_ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c", + "job": "dsh-dev34-install-pair-16-20261030", + "trial": "mteb-leaderboard__KCD6tEf", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-16-20261030.json", + "sha256": "03782151798a8c423bf3dd80c0381749060f14443fa18573990d0a52cece5bc3" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/manifest.json", + "sha256": "757978d6b034758a75027aac61264ae6eb72243da9338ffafdeae28cfa78051b" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030-input/summary.json", + "sha256": "40055254b4c8127df523ee89e946b0426f1f65b2935643c17bb447d5f40f3a5c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/result.json", + "sha256": "06d54613223f16144e4d17d9c12e0a9061487bc4fffba4d0d146ef0e55cefdfb" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-16-20261030/mteb-leaderboard__KCD6tEf/agent/bootstrap.json", + "sha256": "8a30f5515bd2ccd7749dc2993f8a83b0b668c109ec78febbf08cad59e3b4caa1" + } + ] + }, + { + "task": "terminal-bench/build-pov-ray", + "task_ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb", + "job": "dsh-dev34-install-pair-17-20261030", + "trial": "build-pov-ray__TbS8MbA", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-17-20261030.json", + "sha256": "a0b54e31b9726f209c2972852b2ed83fc504600eace0db69d14235ef16e91e0a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/manifest.json", + "sha256": "f8e99d12bdfa6699177b64203869a4cee08d713a5349a251f445a60eb6eddd80" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/summary.json", + "sha256": "32a7d803060d971c6db5bfe036524455d3000efa4997f37f4e8887acc1d62e32" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/result.json", + "sha256": "f9a470171cc88e6a3728a04b99d9ef0d9abf92d9768cc9e63cbdd93686b1582c" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/build-pov-ray__TbS8MbA/agent/bootstrap.json", + "sha256": "0256d0d014edd0ec4cde0e209dc02e20f807b269ad243938f203c3120b85bc9d" + } + ] + }, + { + "task": "terminal-bench/regex-chess", + "task_ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78", + "job": "dsh-dev34-install-pair-17-20261030", + "trial": "regex-chess__3FtpzJU", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": null, + "status": "not_configured" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-install-pair-17-20261030.json", + "sha256": "a0b54e31b9726f209c2972852b2ed83fc504600eace0db69d14235ef16e91e0a" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/manifest.json", + "sha256": "f8e99d12bdfa6699177b64203869a4cee08d713a5349a251f445a60eb6eddd80" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030-input/summary.json", + "sha256": "32a7d803060d971c6db5bfe036524455d3000efa4997f37f4e8887acc1d62e32" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/result.json", + "sha256": "a0d6f78c049b56f4ca2e954e982800a488134b1c3b3af8b1fbe10abc701e386d" + }, + { + "path": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-install-pair-17-20261030/regex-chess__3FtpzJU/agent/bootstrap.json", + "sha256": "62d476124327f88ea2dcd763179ac4379f3c0bb29655570666f2daa5437e8c86" + } + ] + } + ], + "image_transport_audit": { + "path": "jobs/deepseek-harness-dev34-20261030-orchestration/image-cache.jsonl", + "sha256": "523c0f267f8523ebb27d8e8fd0f0743811d6ebf4ef2da113cd654b75aae203fa", + "actions": { + "restored": 56, + "verified": 68, + "evicted": 56 + }, + "policy": "Original image identities and repository digests checked for every preflight/formal task. Only newly introduced image references evicted without force; initially protected images retained." + }, + "exposure_audit": { + "tool_calls_scanned": 675, + "method": "After freezing all outputs, targeted scan of recorded tool calls and paired candidate outputs for benchmark repositories, hidden test paths and environment access; exact byte matching of the authorized provider credential across 2506 frozen files. Not an exhaustive proof of absence of exposure or exfiltration.", + "benchmark_material_observed": false, + "findings": "Terminal-Bench strings in a calendar PRODID and the task-provided C++ canary are not reference retrieval. MTEB accessed legitimate public embeddings-benchmark/results application data at 3125e8302256ac01d4dd68b6f92902b3e6a78987. No external Terminal-Bench reference retrieval or official hidden test access was observed in the scanned calls and candidate outputs.", + "credential_exact_match_files": 0, + "raw_trajectories_published": false + }, + "sdk_lifecycle_diagnostic": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json", + "sha256": "4257a96a3415a7d389fb7b3fb10f819341714ba833b0911eb2a6468171e70c47" + }, + "supporting_evidence": [ + { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/setup-evidence.json", + "sha256": "118edc62f44e571ee2df5bbd41f6ab6fffc78a22908869dcaeffd11c6c9a8fbf" + }, + { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/preflight-evidence.json", + "sha256": "06a724f349f7ddf43cff4983e203dc192a39dd08b1e21f561bfce906e90de3e9" + }, + { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/profile-metadata-amendment.json", + "sha256": "f27cf8b8939506f1ec68d964322aa9ef4af6da4a0f2e9f95c1b32aaed59fb7de" + }, + { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/launch-evidence.json", + "sha256": "c520696465ca9b64ee46e2fb786945abb9043f74baf0851c828404781d2d7cf3" + }, + { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/model-alias-check.json", + "sha256": "cb6be7e927653fa7c7bef9110262dbe9fef6ea75db65964c04a7aa3026c1c9bb" + }, + { + "path": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/deepseek-distribution.json", + "sha256": "34410bdce15ec3b417105eb7f9f6e01a14c47f1bfd0ffe16d76338e7efa51dd2" + } + ], + "trials": [ + { + "task": "terminal-bench/adaptive-rejection-sampler", + "task_ref": "sha256:bcaa2399985cd57666018025846289ab25e193ae0dd8fb7f0ffab2410c24d4de", + "job_name": "dsh-dev34-adaptive-rejection-sampler-20261030", + "trial": "adaptive-rejection-sampler__YPNqHNb", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-adaptive-rejection-sampler-20261030.json", + "native_reward": 0.0, + "native_exception_type": "AgentTimeoutError", + "standard_summary_status": "execution_error", + "standard_summary_rewards": null, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "cancelled", + "processes_stopped": true, + "stopped_processes": 10 + }, + "sdk_finish_reason": null, + "pytest": { + "passed": 4, + "failed": 5, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 9, + "passed": 4, + "failed": 5, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793447384.2657382, + "stop": 1793447384.7044072 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "timeout", + "duration_seconds": 900.002259612 + }, + { + "attempt": 2, + "status": "timeout", + "duration_seconds": 900.0019231870001 + }, + { + "attempt": 3, + "status": "completed", + "return_code": 0, + "duration_seconds": 26.260310242000287 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:03:16.295512Z", + "finished_at": "2026-10-31T11:03:18.479964Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:03:18.480032Z", + "finished_at": "2026-10-31T11:04:18.589676Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:04:18.589881Z", + "finished_at": "2026-10-31T11:19:18.029412Z" + }, + "verifier": { + "started_at": "2026-10-31T11:19:18.279387Z", + "finished_at": "2026-10-31T11:49:44.806796Z" + } + }, + "agent_seconds": 899.439531, + "usage": { + "n_input_tokens": 257358, + "n_output_tokens": 38506, + "n_cache_tokens": 249216, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 13, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": true, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 13, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 13, + "assistant/message": 13, + "tool/call": 13, + "tool/result": 12, + "step/end": 12 + }, + "event_count": 85, + "tool_calls": 13, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": null, + "classification": "unscored_invalid_boundary_and_verifier_error", + "reference_exposure": false, + "agent_seconds": 901.286359 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-adaptive-rejection-sampler-20261030.json": "07e155e9a7eb9063c06d0786611f980a43720128c06f0027c900b835d36c7d25", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/result.json": "91fb7759d91c13281d4753262b4f9d41e902b877d9ccb286980e3ea5263d0b76", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/manifest.json": "e5cb361bfbde60f4ebcd3cae096558dd7d60115c43089163450eb9d2fff24260", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030-input/config.json": "a706537d7c72f1b91b34c58ed0e35496240e66329069eab0ee8ae64e7717653f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/process-supervision.json": "9fd361bfccc68573302510b243fe0f8da598f5fc1449cc84b92d96fb56b87c41", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/agent/dsh-events.jsonl": "fa697cfae5ae7fee611834c6c2147bebdb7c7338c6f8d24f19c93442d646a168", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/test-stdout.txt": "923bae414e71ab4aa5ac4666eb5342bd2ccb94dc547f1ad4b96c250ec31a564b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-adaptive-rejection-sampler-20261030/adaptive-rejection-sampler__YPNqHNb/verifier/retry-summary.json": "2713972eb5f28c99c93cc7137ffc023fe6a9be9b49c680386d0960c21484cd83" + }, + "diagnosis": "Native 900-second agent limit; cancellation confirmed with 10 stopped processes. Final official pytest ran after two verifier timeouts: Rscript and generated sample files were missing (4 passed, 5 failed). Installing R when absent is an explicit task requirement, so this is a scored task failure.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/build-cython-ext", + "task_ref": "sha256:35dc561790e73f13e59837828ac0863c0893edc6b71d42cf7d58656c843adf12", + "job_name": "dsh-dev34-build-cython-ext-20261030", + "trial": "build-cython-ext__vRLUmiS", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-cython-ext-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 11, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 11, + "passed": 11, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793445244.0513752, + "stop": 1793445264.69225 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 29.697184281999967 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:03:19.170345Z", + "finished_at": "2026-10-31T11:03:21.297508Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:03:21.297549Z", + "finished_at": "2026-10-31T11:04:00.792658Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:04:00.792847Z", + "finished_at": "2026-10-31T11:13:53.883037Z" + }, + "verifier": { + "started_at": "2026-10-31T11:13:54.109770Z", + "finished_at": "2026-10-31T11:14:24.932637Z" + } + }, + "agent_seconds": 593.09019, + "usage": { + "n_input_tokens": 1566507, + "n_output_tokens": 12153, + "n_cache_tokens": 1516672, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 48, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 48, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 48, + "assistant/message": 48, + "tool/call": 47, + "tool/result": 47, + "step/end": 48, + "turn/end": 1 + }, + "event_count": 296, + "tool_calls": 47, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 834.194182 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-cython-ext-20261030.json": "7399a5504eb1875249ebb046eb0008c52c5159e3a1296b1e1893a9be194cf41d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/result.json": "a11acf668fc25f24e0504f5435a9872401145fc601ac94645f9dab6e41704725", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/manifest.json": "e5cb361bfbde60f4ebcd3cae096558dd7d60115c43089163450eb9d2fff24260", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030-input/config.json": "aa3f055b4e578b91869d2d6cac27ff50ea8f1a5f597ff0b3a770f64b21ac6a2e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/agent/dsh-events.jsonl": "6f96b40fb8f18b878b20bb161e816601b4f7c2197b0409f3a583ea0fd1d3f7bc", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/test-stdout.txt": "9de6887f99b6567a850f9455ee552bd5f63021d34a9f03e4979c0aff98cb4ed3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-cython-ext-20261030/build-cython-ext__vRLUmiS/verifier/retry-summary.json": "1320ddee25ebcdc4bd943a4141e9e19b61116e3ae973b9166f889a73b559c1a0" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/cancel-async-tasks", + "task_ref": "sha256:a3d048d351136e48070696cda8bb79660dfd74db1fea3b6da88559f0332699c1", + "job_name": "dsh-dev34-cancel-async-tasks-20261030", + "trial": "cancel-async-tasks__ZM2fotf", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cancel-async-tasks-20261030.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 5, + "failed": 1, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 5, + "failed": 1, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793447498.2467968, + "stop": 1793447510.9919577 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 29.119335383999896 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:50:23.989272Z", + "finished_at": "2026-10-31T11:50:26.082120Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:50:26.082169Z", + "finished_at": "2026-10-31T11:50:55.136884Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:50:55.137025Z", + "finished_at": "2026-10-31T11:51:21.481018Z" + }, + "verifier": { + "started_at": "2026-10-31T11:51:21.714141Z", + "finished_at": "2026-10-31T11:51:51.059411Z" + } + }, + "agent_seconds": 26.343993, + "usage": { + "n_input_tokens": 9214, + "n_output_tokens": 2918, + "n_cache_tokens": 7808, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 4, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 4, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 4, + "assistant/message": 4, + "tool/call": 3, + "tool/result": 3, + "step/end": 4, + "turn/end": 1 + }, + "event_count": 32, + "tool_calls": 3, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 104.775936 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cancel-async-tasks-20261030.json": "0f64ab7b7f433bc1dd9a742f90871e54d7b9da64d3faeb74dcd6676a6956553c", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/result.json": "ab0c8d2685ec1064d386684942774d41c013ad649dbc78a1b9f4a6e2e3b2d3e9", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/manifest.json": "deff376d7643827846095b81ae44f57092fdca37913c59372d09fce0f72b35c7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030-input/config.json": "9d21a9336976fe82458b423ad8bd14979a087d1badacea94e5570bf97e94152b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/agent/dsh-events.jsonl": "9e7945012332239fc12fd55667a630ff199e857a73a0a63ee71df31a282e1ae3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/test-stdout.txt": "f0a8804f6456d9a566e33de1bb448baeebce37177b8474f9b99401ebf4d64fda", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cancel-async-tasks-20261030/cancel-async-tasks__ZM2fotf/verifier/retry-summary.json": "ba3ea17b9be77ffc531fb345c98ed7cbd38c57aaed9c202bc6c3b6014cd37c95" + }, + "diagnosis": "Official assertion found zero excess-task cleanup messages, expected two (5 passed, 1 failed).", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/cobol-modernization", + "task_ref": "sha256:e03aa03965eee48c9f772d566ff9bcd741fceeac9693123f82a13fdd46be0d88", + "job_name": "dsh-dev34-cobol-modernization-20261030", + "trial": "cobol-modernization__4f7KNiZ", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cobol-modernization-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 3, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793447641.4687269, + "stop": 1793447641.5622828 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 15.844012083999587 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:50:24.998779Z", + "finished_at": "2026-10-31T11:50:27.070139Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:50:27.070212Z", + "finished_at": "2026-10-31T11:50:52.999173Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:50:52.999394Z", + "finished_at": "2026-10-31T11:53:45.357599Z" + }, + "verifier": { + "started_at": "2026-10-31T11:53:45.583177Z", + "finished_at": "2026-10-31T11:54:01.615769Z" + } + }, + "agent_seconds": 172.358205, + "usage": { + "n_input_tokens": 295337, + "n_output_tokens": 19905, + "n_cache_tokens": 283776, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 20, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 20, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 20, + "assistant/message": 20, + "tool/call": 21, + "tool/result": 21, + "step/end": 20, + "turn/end": 1 + }, + "event_count": 132, + "tool_calls": 21, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 430.243181 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-cobol-modernization-20261030.json": "e2eaf39089534ae6b42cfb53e2af57b0efc1e910f8ce3100cd24be07039c2f55", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/result.json": "548dd8306a610d37e299f879e648ed69405af406346231be18f0c34a680092c8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/manifest.json": "deff376d7643827846095b81ae44f57092fdca37913c59372d09fce0f72b35c7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030-input/config.json": "c8c45654330a62fbbee86de53236aa014930b6722b58677d8c5beed4b4d561ff", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/agent/dsh-events.jsonl": "53339f995859d05413b8aee55ce34f7b46871518ceca70d1385a39f4f6269c46", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/test-stdout.txt": "ef4690316660e8cef4af73912d0b4140e3f96212795cc4770bcfbab33eea1289", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-cobol-modernization-20261030/cobol-modernization__4f7KNiZ/verifier/retry-summary.json": "b2ea13ea8639e2f51a9d0d4a7e1150301d022cb83f09589db8a04ed8581a5e0a" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/db-wal-recovery", + "task_ref": "sha256:07e9193513bfce038b9f9d84962ecc85683eee0a642d14ee6fb7104b086ddab4", + "job_name": "dsh-dev34-db-wal-recovery-20261030", + "trial": "db-wal-recovery__B45tM5E", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-db-wal-recovery-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 7, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 7, + "passed": 7, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793447750.0996993, + "stop": 1793447750.145449 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 18.101348572000006 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:54:30.096881Z", + "finished_at": "2026-10-31T11:54:32.177051Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:54:32.177094Z", + "finished_at": "2026-10-31T11:54:57.568379Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:54:57.568571Z", + "finished_at": "2026-10-31T11:55:31.657414Z" + }, + "verifier": { + "started_at": "2026-10-31T11:55:31.900442Z", + "finished_at": "2026-10-31T11:55:50.224387Z" + } + }, + "agent_seconds": 34.088843, + "usage": { + "n_input_tokens": 61777, + "n_output_tokens": 2565, + "n_cache_tokens": 50688, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 7, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 7, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 7, + "assistant/message": 7, + "tool/call": 6, + "tool/result": 6, + "step/end": 7, + "turn/end": 1 + }, + "event_count": 50, + "tool_calls": 6, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": null, + "classification": "unscored_unverified_timeout_boundary", + "reference_exposure": false, + "agent_seconds": 898.690758 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-db-wal-recovery-20261030.json": "26244bc83c6648aa9de4760f5446ec55e2491726ca8150fd1374d6159204c232", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/result.json": "84708abe150c2fcbb37e5dd8449d5f950dc4f6b23a23e49859c7d7f70d643d18", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/manifest.json": "1b9a6fcc70ee8a059b5cfb1d9a1af34a34798426c27264c25924c43ceb8b1031", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030-input/config.json": "14de9939fccf109fc585ee64b829d37acd26382e43b8f8559c926fc7afdc89a4", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/agent/dsh-events.jsonl": "b74c7f1e760eb5e2ee7ad1ad323c4b0280c5b2d55e1c0007059a667a5efba7b2", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/test-stdout.txt": "0a393eb66bc19a31f0f070f2ea377d8b0e11cb35b8130f44155a59249239d546", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-db-wal-recovery-20261030/db-wal-recovery__B45tM5E/verifier/retry-summary.json": "253bc8ac712beccdc4eae0b28c56c5ad6d00f723ee173223d5ccddee6edeac8b" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/fix-code-vulnerability", + "task_ref": "sha256:d31348aa16b533a15f013420e4d9726dc529e6086adc7d6d48067d22cc18fe71", + "job_name": "dsh-dev34-fix-code-vulnerability-20261030", + "trial": "fix-code-vulnerability__UBxed3u", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-fix-code-vulnerability-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 6, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 6, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793447749.986513, + "stop": 1793447750.0051932 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 9.682398951999858 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:54:29.935241Z", + "finished_at": "2026-10-31T11:54:32.135299Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:54:32.135352Z", + "finished_at": "2026-10-31T11:54:57.937537Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:54:57.937743Z", + "finished_at": "2026-10-31T11:55:40.002902Z" + }, + "verifier": { + "started_at": "2026-10-31T11:55:40.231645Z", + "finished_at": "2026-10-31T11:55:50.110489Z" + } + }, + "agent_seconds": 42.065159, + "usage": { + "n_input_tokens": 87921, + "n_output_tokens": 2646, + "n_cache_tokens": 74496, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 11, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 11, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 11, + "assistant/message": 11, + "tool/call": 10, + "tool/result": 10, + "step/end": 11, + "turn/end": 1 + }, + "event_count": 74, + "tool_calls": 10, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 28.232551 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-fix-code-vulnerability-20261030.json": "76f1d136fddd04cf4fd5aaa913de183569eeded32018af958bd49e11de40d6bd", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/result.json": "26045760b379e3ea79996937a90daf2e0ae44214cd0a8e3c0f0e2fc514ccc7ca", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/manifest.json": "1b9a6fcc70ee8a059b5cfb1d9a1af34a34798426c27264c25924c43ceb8b1031", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030-input/config.json": "95fb2f6329d9f71e31ae9ddb1cc889e873d8e466dfc2043581afc2d22f04be07", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/agent/dsh-events.jsonl": "9053d5ec1e3f44061030b500717b4565bf2bf7917932a3c28accda4b73b8dd27", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/test-stdout.txt": "bde9f138d8ab9eb7a0e534edd3f4e2a68ce7ec2916e1f32efcc61cbcf6867953", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-fix-code-vulnerability-20261030/fix-code-vulnerability__UBxed3u/verifier/retry-summary.json": "6e307bbd5557dc4dc6f55f6b7aa719ee94adc1292a1471970de1e41a14332cf3" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/git-multibranch", + "task_ref": "sha256:7759a92a373c967dd31535975bad8659410644025673fe62385444fd2605c5fa", + "job_name": "dsh-dev34-git-multibranch-20261030", + "trial": "git-multibranch__oB5UP5P", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-git-multibranch-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 1, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 1, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793447957.4885902, + "stop": 1793447998.4306827 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 93.5592820830002 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:56:33.599077Z", + "finished_at": "2026-10-31T11:56:35.650025Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:56:35.650085Z", + "finished_at": "2026-10-31T11:57:01.001067Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:57:01.001183Z", + "finished_at": "2026-10-31T11:58:24.527992Z" + }, + "verifier": { + "started_at": "2026-10-31T11:58:24.752678Z", + "finished_at": "2026-10-31T11:59:58.506882Z" + } + }, + "agent_seconds": 83.526809, + "usage": { + "n_input_tokens": 170900, + "n_output_tokens": 8458, + "n_cache_tokens": 161408, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 16, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 16, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 16, + "assistant/message": 16, + "tool/call": 15, + "tool/result": 15, + "step/end": 16, + "turn/end": 1 + }, + "event_count": 104, + "tool_calls": 15, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 80.488897 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-git-multibranch-20261030.json": "9d5aa4fcd1e0a8c742324a51a6535128cbf4a159a44451f10406f60be7b7362f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/result.json": "ebb44af0bea23aeca0f330db173e94063d5c2a0c7aac04df51ea5a01f7a2d91e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/manifest.json": "2f30ce0ec6c9ed8b8bd07f8bbaa530c5ef13746d58639b7641d9084d318e916f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030-input/config.json": "f5663bea26f53ae2d68148781b108897e80139f774627c332f9429f84d776a08", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/agent/dsh-events.jsonl": "4ad88ed8bac1c2b85bf361bfa47a1d7e5b8465aa7396f27246b2f637b091949c", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/test-stdout.txt": "143fb90bbccce960c0dd95dad060fc95ac6ff18b55860c324222cfc86f62dfb6", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-git-multibranch-20261030/git-multibranch__oB5UP5P/verifier/retry-summary.json": "30af07961a220d1d67ea80695c193564e627a58f47e7fde3fe24dff059c137cb" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/kv-store-grpc", + "task_ref": "sha256:973c5d4c111fb61a344457936f1c36400acd2d9e44389e7b319586fe23a7a307", + "job_name": "dsh-dev34-kv-store-grpc-20261030", + "trial": "kv-store-grpc__5c55ojh", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-kv-store-grpc-20261030.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 5, + "failed": 2, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 7, + "passed": 5, + "failed": 2, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793448067.3066227, + "stop": 1793448067.3952668 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 195.3415425889998 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T11:56:33.286330Z", + "finished_at": "2026-10-31T11:56:35.430293Z" + }, + "agent_setup": { + "started_at": "2026-10-31T11:56:35.430351Z", + "finished_at": "2026-10-31T11:57:05.872769Z" + }, + "agent_execution": { + "started_at": "2026-10-31T11:57:05.872981Z", + "finished_at": "2026-10-31T11:57:51.690437Z" + }, + "verifier": { + "started_at": "2026-10-31T11:57:51.924784Z", + "finished_at": "2026-10-31T12:01:07.472069Z" + } + }, + "agent_seconds": 45.817456, + "usage": { + "n_input_tokens": 13266, + "n_output_tokens": 1640, + "n_cache_tokens": 11008, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 6, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 6, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 6, + "assistant/message": 6, + "tool/call": 5, + "tool/result": 5, + "step/end": 6, + "turn/end": 1 + }, + "event_count": 44, + "tool_calls": 5, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 36.133692 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-kv-store-grpc-20261030.json": "be9932537f63a21db0eeba1dd769a381ed12d7a4062fe46409a0de6b39075852", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/result.json": "aa9e6b5d4ac06b90685252efbb142dc8047f6c9f7b764138c39e6028541e0e35", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/manifest.json": "2b8f9822250021cd497f3536b485da8c104e6cca9d80a49ddde7820745f8deaa", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030-input/config.json": "e0b670a59da7c8751fa10f2489c0730b4a1c89f1d49479f14de51d76ae4391b3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/agent/dsh-events.jsonl": "24ac5c6765b18ad9b0598642b3a13ac7edb05f40ce7fc9bc18ba965a2a08de49", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/test-stdout.txt": "7fa5fc2c891910c002d4cf8836dd23b2b8e44fcae77018c6c4f9c9d9a4163de4", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-kv-store-grpc-20261030/kv-store-grpc__5c55ojh/verifier/retry-summary.json": "7b1e40b2c71890f9e38e3595c850e13371db3253622d3c6112b47dbdc3796108", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json": "4257a96a3415a7d389fb7b3fb10f819341714ba833b0911eb2a6468171e70c47" + }, + "diagnosis": "Official verification found port 5328 not listening and connection refused (5 passed, 2 failed). Agent self-tests had successfully called SetVal/GetVal with value 123 before SDK shutdown. The post-freeze probe confirms normal SDK closure terminates a background server. Retain the measured failure with a service-lifecycle caveat; do not infer incorrect RPC implementation from this result.", + "audit_flags": { + "service_lifecycle_caveat": true, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/log-summary-date-ranges", + "task_ref": "sha256:27b074a2f10fff7606e096f3abd8dced418ad8fda0f53d88acbe477f2d9ceaf6", + "job_name": "dsh-dev34-log-summary-date-ranges-20261030", + "trial": "log-summary-date-ranges__2svaJZY", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-log-summary-date-ranges-20261030.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 1, + "failed": 1, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 2, + "passed": 1, + "failed": 1, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793448210.4891813, + "stop": 1793448210.5345502 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 38.036719163999805 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:01:53.341381Z", + "finished_at": "2026-10-31T12:01:55.435363Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:01:55.435405Z", + "finished_at": "2026-10-31T12:02:31.185634Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:02:31.185757Z", + "finished_at": "2026-10-31T12:02:52.148482Z" + }, + "verifier": { + "started_at": "2026-10-31T12:02:52.380236Z", + "finished_at": "2026-10-31T12:03:30.610850Z" + } + }, + "agent_seconds": 20.962725, + "usage": { + "n_input_tokens": 28124, + "n_output_tokens": 1744, + "n_cache_tokens": 19968, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 4, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 4, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 4, + "assistant/message": 4, + "tool/call": 3, + "tool/result": 3, + "step/end": 4, + "turn/end": 1 + }, + "event_count": 32, + "tool_calls": 3, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 13.883248 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-log-summary-date-ranges-20261030.json": "31de4194d1734a7210a35e87cfc0336575d0dc61e7ac1b2b7e04cbaa2537e075", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/result.json": "d02b44a1b25ef6fa02f8986f5bcebee4d587b77b6a74f8dfb1b95dba292211fb", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/manifest.json": "fcb2bfde554aea943ae7ea256b857fde7e0f18a29d452fa95289a246e474fd52", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030-input/config.json": "94178d924b996009ef571fa6bc69301551be952b3bae9ac66175208ec0a41aa7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/agent/dsh-events.jsonl": "4030669235c95d1af8ae4ffb456e23480867f9e30e81fe480e58a89b13100377", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/test-stdout.txt": "d9446e5503fc0cea5ca1b2b061c0bc84c9a378a1d4c34ab8fc59666285b575e7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-log-summary-date-ranges-20261030/log-summary-date-ranges__2svaJZY/verifier/retry-summary.json": "ffd57a60e6e19e4bf54d33c853f9010a43030bef21612dc1d44fe81433c43e77" + }, + "diagnosis": "Official assertion found today ERROR count 74, expected 370 (1 passed, 1 failed).", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/merge-diff-arc-agi-task", + "task_ref": "sha256:6aab6511a5344ce87698293bb1ce4cc51d9a45f1ad9f0c075d2a83197b36727d", + "job_name": "dsh-dev34-merge-diff-arc-agi-task-20261030", + "trial": "merge-diff-arc-agi-task__7r5AC2q", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-merge-diff-arc-agi-task-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 5, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 5, + "passed": 5, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793448291.23935, + "stop": 1793448291.2812428 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 26.77169537000009 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:01:51.900043Z", + "finished_at": "2026-10-31T12:01:54.138809Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:01:54.138853Z", + "finished_at": "2026-10-31T12:02:34.299271Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:02:34.299462Z", + "finished_at": "2026-10-31T12:04:24.760970Z" + }, + "verifier": { + "started_at": "2026-10-31T12:04:24.988511Z", + "finished_at": "2026-10-31T12:04:51.382784Z" + } + }, + "agent_seconds": 110.461508, + "usage": { + "n_input_tokens": 104072, + "n_output_tokens": 12323, + "n_cache_tokens": 98304, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 13, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 13, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 13, + "assistant/message": 13, + "tool/call": 12, + "tool/result": 12, + "step/end": 13, + "turn/end": 1 + }, + "event_count": 86, + "tool_calls": 12, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 77.083453 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-merge-diff-arc-agi-task-20261030.json": "702a32470ded74f31aac195113a429e265de167beec4bc0d0105c516cedc5bea", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/result.json": "a863044a3edb62310b85043a6c9cbe76f907427566b4fa463e6b77dc8bd7ab11", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/manifest.json": "fcb2bfde554aea943ae7ea256b857fde7e0f18a29d452fa95289a246e474fd52", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030-input/config.json": "1ae54d8b5c7de6aa5e58cf9974b4a3747b895de05951dcee7f21134b0685063f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/agent/dsh-events.jsonl": "746fb7b6120b33b3ce2d6225c56fa9738a634586bc441f0abd9f08b949a55490", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/test-stdout.txt": "71ed332ca5614ceb6b4267447b80d71a048b7d55a5f50e5a47b987d84ba8732f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-merge-diff-arc-agi-task-20261030/merge-diff-arc-agi-task__7r5AC2q/verifier/retry-summary.json": "e41419110d4702693b481e9f76ce2eede341dbdf7c9deb84fd3745eaa03b3163" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/model-extraction-relu-logits", + "task_ref": "sha256:1ae5045ad68b5d34c3398b612066a07c4a08b6dc330d28868ec4021e17c94b17", + "job_name": "dsh-dev34-model-extraction-relu-logits-20261030", + "trial": "model-extraction-relu-logits__QcM84FH", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-model-extraction-relu-logits-20261030.json", + "native_reward": 0.0, + "native_exception_type": "AgentTimeoutError", + "standard_summary_status": "execution_error", + "standard_summary_rewards": null, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "cancelled", + "processes_stopped": true, + "stopped_processes": 4 + }, + "sdk_finish_reason": null, + "pytest": { + "passed": 0, + "failed": 1, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 0, + "failed": 1, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793449388.9962518, + "stop": 1793449389.3775601 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 116.58492313500028 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:05:32.657753Z", + "finished_at": "2026-10-31T12:05:34.756233Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:05:34.756283Z", + "finished_at": "2026-10-31T12:06:12.649900Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:06:12.650028Z", + "finished_at": "2026-10-31T12:21:12.003257Z" + }, + "verifier": { + "started_at": "2026-10-31T12:21:12.230894Z", + "finished_at": "2026-10-31T12:23:09.456413Z" + } + }, + "agent_seconds": 899.353229, + "usage": { + "n_input_tokens": 874923, + "n_output_tokens": 46751, + "n_cache_tokens": 848384, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 22, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": true, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 23, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 22, + "assistant/message": 22, + "tool/call": 22, + "tool/result": 22, + "step/end": 22 + }, + "event_count": 142, + "tool_calls": 22, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 176.964303 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-model-extraction-relu-logits-20261030.json": "bd98b06439274f90039f45dd6707fb61cb4f58f02e12b4e9066054d480912d3a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/result.json": "e934b2adf693122b884894db9146c87ba83224b684a05c3d5a081d9f40c2e855", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/manifest.json": "a9394befa66cbac76b4ae051b0f95c936e12d7c7aa628e1b094927bd47895358", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030-input/config.json": "489009bd9f3164fc507508ffdc3c2e5ed87b48ac766019716cdeb4e444f81711", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/process-supervision.json": "ecf9ddfe1194866091d36f7817a57689da8cdd81ee2df92be1c550c2f3edddd3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/agent/dsh-events.jsonl": "a4bfa78a56640ebc50e51ef50d16f6ff84be22f9edca084554d85e3a6503ffa0", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/test-stdout.txt": "5b663727ff352b125abe6e2ce9219cd9e086c78dd203754b7aed5f27e132affb", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-model-extraction-relu-logits-20261030/model-extraction-relu-logits__QcM84FH/verifier/retry-summary.json": "05d636ce9058e41247dd69cacbc748590b45eaea1cd509e40e88ede6c6048a83" + }, + "diagnosis": "Native 900-second agent limit; cancellation confirmed with 4 stopped processes. Official verification could not find /app/stolen_A1.npy (0 passed, 1 failed).", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/multi-source-data-merger", + "task_ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "job_name": "dsh-dev34-multi-source-data-merger-20261030", + "trial": "multi-source-data-merger__4guMJdj", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-multi-source-data-merger-20261030.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "unscored_verifier_error", + "analytical_score": null, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": null, + "ctrf_summary": null, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 349.03329265200045 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:05:31.534919Z", + "finished_at": "2026-10-31T12:05:33.733657Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:05:33.733699Z", + "finished_at": "2026-10-31T12:06:12.508680Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:06:12.508808Z", + "finished_at": "2026-10-31T12:06:40.340730Z" + }, + "verifier": { + "started_at": "2026-10-31T12:06:40.576506Z", + "finished_at": "2026-10-31T12:12:29.971691Z" + } + }, + "agent_seconds": 27.831922, + "usage": { + "n_input_tokens": 8625, + "n_output_tokens": 3255, + "n_cache_tokens": 6400, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 3, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 3, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 3, + "assistant/message": 3, + "tool/call": 2, + "tool/result": 2, + "step/end": 3, + "turn/end": 1 + }, + "event_count": 26, + "tool_calls": 2, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 34.336406 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-multi-source-data-merger-20261030.json": "38112c9c1650892699c571ef0d97074725b0f4fa0f1e1711340acc05f8ee8857", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/result.json": "8d3dbf7742c155f83ff42673a35fbdbf336cc76e6014e7b44425f3aa36464a56", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/manifest.json": "6fa7bc807d925fbd687aca99feda437bfd33a0b1daf801b71266b0206e32692b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030-input/config.json": "8e8d48ac943493b44c102825fcea323bd6ec9b000a09d2370878278789f4b518", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh-events.jsonl": "5952cc8a6ab085f0160b13e29fa868f47827d01bb175ef51b1e66c017fee3b3a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/test-stdout.txt": "790fab06d71b6afc57e26aff4a9ff7da1df45708a2dc0abb5b6c1a4809d7c3d5", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/verifier/retry-summary.json": "aaa5241bdfee3617dc074356809311cab05e39bfc565bcfcf6c648b7e6ea377b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/trial.log": "0b70cbdaf5a0e27bd10c89eb37657b1453cf2c28cb6d36efba6de1746783674d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/dsh-result.json": "e11b06159d9a0379f5d3997cbf4a2a6d8630bd441f8272eec1e08740ff2cb2eb", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-multi-source-data-merger-20261030/multi-source-data-merger__4guMJdj/agent/bootstrap.json": "38a018af813827b21bff5685e056300a44dc7899ae803e02ae564568808be3fd" + }, + "diagnosis": "Official verifier failed downloading/extracting pyarrow==22.0.0 with UV_HTTP_TIMEOUT=30 before pytest began. Native reward zero retained; analytical score null.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/nginx-request-logging", + "task_ref": "sha256:9d1b8bebd989ea0bc8080c3b159480068caf183c5fd61385868b2574b206e097", + "job_name": "dsh-dev34-nginx-request-logging-20261030", + "trial": "nginx-request-logging__Zd8i978", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-nginx-request-logging-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 8, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 8, + "passed": 8, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793449661.346323, + "stop": 1793449663.5238392 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 44.802392953999515 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:23:52.836911Z", + "finished_at": "2026-10-31T12:23:54.990238Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:23:54.990302Z", + "finished_at": "2026-10-31T12:24:40.438824Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:24:40.439014Z", + "finished_at": "2026-10-31T12:26:58.409400Z" + }, + "verifier": { + "started_at": "2026-10-31T12:26:58.630908Z", + "finished_at": "2026-10-31T12:27:43.640105Z" + } + }, + "agent_seconds": 137.970386, + "usage": { + "n_input_tokens": 19910, + "n_output_tokens": 2739, + "n_cache_tokens": 16384, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 6, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 6, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 6, + "assistant/message": 6, + "tool/call": 5, + "tool/result": 5, + "step/end": 6, + "turn/end": 1 + }, + "event_count": 44, + "tool_calls": 5, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 52.085658 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-nginx-request-logging-20261030.json": "b7334855e18b0beb942d1c6a8fbae1b60a540377a57ec00e720a0f6ea6b032c0", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/result.json": "d51ec031edb07f74bcaf32527fda2fe19a88526868709c194be3740fb1e84add", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/manifest.json": "39b824fc7941d63fcb1db8e9cd0f9bf7ffe6b852de5200e9d03240bbb409a20a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030-input/config.json": "f62567d789a0dce0e0ae922b869bcce1bdd64cbd5d78d1f9479fef47cc5ccf78", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/agent/dsh-events.jsonl": "366acd57bcddd66e8415c0b9e6dc2d3bcb4cd40e755f9668de001a1cda0ab166", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/test-stdout.txt": "fc0078ba84a3973a5b78789011cad03edfd96c4485e85f54b4869f03865055d8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-nginx-request-logging-20261030/nginx-request-logging__Zd8i978/verifier/retry-summary.json": "82eb40e1ac90262fd6a9ac5f1401a3df82724c2f47522c2361867acba073a562" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/openssl-selfsigned-cert", + "task_ref": "sha256:d4afa2bd2a9ba1420db8d6cfde42ffdb4873ae2d955c35014e8da94444c83302", + "job_name": "dsh-dev34-openssl-selfsigned-cert-20261030", + "trial": "openssl-selfsigned-cert__CQmHcJS", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-openssl-selfsigned-cert-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 6, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 6, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793449577.673935, + "stop": 1793449577.8550599 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 66.24086833199999 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:23:53.930120Z", + "finished_at": "2026-10-31T12:23:56.026842Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:23:56.026885Z", + "finished_at": "2026-10-31T12:24:35.064484Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:24:35.064671Z", + "finished_at": "2026-10-31T12:25:11.236111Z" + }, + "verifier": { + "started_at": "2026-10-31T12:25:11.479789Z", + "finished_at": "2026-10-31T12:26:17.915199Z" + } + }, + "agent_seconds": 36.17144, + "usage": { + "n_input_tokens": 22963, + "n_output_tokens": 3079, + "n_cache_tokens": 20864, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 8, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 8, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 8, + "assistant/message": 8, + "tool/call": 7, + "tool/result": 7, + "step/end": 8, + "turn/end": 1 + }, + "event_count": 56, + "tool_calls": 7, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 16.533788 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-openssl-selfsigned-cert-20261030.json": "28c5d1ddd3499b7ec8a1df2a7db46bbd60d17d7c2dc6639c06be3c1d982f6dee", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/result.json": "820ee3a59fb00dc5cb75f06dbde360353e85b4c7a11ad3a1ae39f1c51c18209b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/manifest.json": "333842ab7163ba3ae33c680d07fe4fabcab34fa5a4d34d5f1faa468130ee60c8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030-input/config.json": "6ef3c739a0cb8a30615d675df06ed1471cd8c9134a19721742ebd2f5003498f3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/agent/dsh-events.jsonl": "d46c3cc427118b37b2738412974c08b412c3d2b0cfb29ff3634ab235b261426d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/test-stdout.txt": "c3d53945b93c1c4d675d491297c1e810beed0eac7468170cb4fc23369b53d3a7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-openssl-selfsigned-cert-20261030/openssl-selfsigned-cert__CQmHcJS/verifier/retry-summary.json": "bfc5efd40036ee3cb99d7c25261057681e5e31b9955b0d4fcf016fbb263573b7" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/pypi-server", + "task_ref": "sha256:1a1e0542f58e2d3362fec17a9bbb98667717d9a4a3e9a4c8413d3150a4fa0ff1", + "job_name": "dsh-dev34-pypi-server-20261030", + "trial": "pypi-server__bRuyWCd", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pypi-server-20261030.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 0, + "failed": 1, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 0, + "failed": 1, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793449876.1825774, + "stop": 1793449884.532053 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 96.41215718400053 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:28:29.406844Z", + "finished_at": "2026-10-31T12:28:31.505972Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:28:31.506012Z", + "finished_at": "2026-10-31T12:29:07.097358Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:29:07.097550Z", + "finished_at": "2026-10-31T12:29:47.259600Z" + }, + "verifier": { + "started_at": "2026-10-31T12:29:47.486657Z", + "finished_at": "2026-10-31T12:31:24.597343Z" + } + }, + "agent_seconds": 40.16205, + "usage": { + "n_input_tokens": 12933, + "n_output_tokens": 3023, + "n_cache_tokens": 11392, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 5, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 5, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 5, + "assistant/message": 5, + "tool/call": 5, + "tool/result": 5, + "step/end": 5, + "turn/end": 1 + }, + "event_count": 40, + "tool_calls": 5, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 70.323736 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pypi-server-20261030.json": "cfe6f68e746db6cfc5bf7873a49af2d0e3e89ff9591114461d1f225c5175be80", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/result.json": "1f924d58cb240440f7bf6a579ce51d51558b4aa6361c2cd21fe78578a7c6964c", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/manifest.json": "eece49ade321c217abae0b62394afcc0db75a4099be275bae9a9b64d5e492dd8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030-input/config.json": "f5effb8af029240b32cf3b1ccda4ab047f63e23e8d8e85befad3eb95b9eecd6a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/agent/dsh-events.jsonl": "f0a93c257e88808cdb0164d14b4b8aae05314a37fc6f4de6413bb1bc69b0fc06", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/test-stdout.txt": "1e005e0457f5a1517d55adc255df5b56a8a2feaef76331757269b87888f5a587", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pypi-server-20261030/pypi-server__bRuyWCd/verifier/retry-summary.json": "2b3fec6f55cdf29fcc426051dc46e3c4a03e0fc7608c57f73e55308a0d067248", + "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/sdk-lifecycle-probe.json": "4257a96a3415a7d389fb7b3fb10f819341714ba833b0911eb2a6468171e70c47" + }, + "diagnosis": "Official pip installation of vectorops==0.1.0 from localhost:8080 failed because the local index was unreachable (0 passed, 1 failed). The agent had started a nohup server, successfully used curl, and installed and exercised the package from a separate virtual environment. The post-freeze SDK closure probe confirms the service-lifecycle caveat. No counterfactual official pass is claimed.", + "audit_flags": { + "service_lifecycle_caveat": true, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/pytorch-model-cli", + "task_ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "job_name": "dsh-dev34-pytorch-model-cli-20261030", + "trial": "pytorch-model-cli__z4gK8xd", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-cli-20261030.json", + "native_reward": null, + "native_exception_type": "NetworkConnectionError", + "standard_summary_status": "setup_failed", + "standard_summary_rewards": null, + "analytical_classification": "unscored_setup_error", + "analytical_score": null, + "boundary_status": "unconfirmed", + "process_supervision": {}, + "sdk_finish_reason": null, + "pytest": null, + "ctrf_summary": null, + "verifier_attempts": null, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:28:27.026202Z", + "finished_at": "2026-10-31T12:28:29.192692Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:28:29.192755Z", + "finished_at": "2026-10-31T12:29:03.734017Z" + }, + "agent_execution": null, + "verifier": null + }, + "agent_seconds": null, + "usage": { + "n_input_tokens": null, + "n_output_tokens": null, + "n_cache_tokens": null, + "cost_usd": null, + "raw_deepseek_metadata": null + }, + "operational_metadata": { + "event_types": {}, + "event_count": 0, + "tool_calls": 0, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 660.719365 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-cli-20261030.json": "d5e1306947c07e0db6d4d1f5ce7236a9be7cc72f089d4796e233ed5a5e43d721", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/result.json": "1e683e794c84c6f143178ce20ddc7ae962bc41452c31cfee27f9863ac20d1b7f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/manifest.json": "f0bb5f17b161155815fd7d02a6d9104717166dcb2fd65b259d721cfce80dba71", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030-input/config.json": "a02841879dce7c30bc5c91d5019403b808f1882d628fb320c2b85d997bbe162c", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/trial.log": "789f84bba478edd0eb153b082bcf479ee69a4f012294f1971661ead5b2c00d47", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/exception.txt": "67cc71bef105517fe1690789812e418d8b27d4aaa36980cb827dcabc55de159a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-cli-20261030/pytorch-model-cli__z4gK8xd/agent/bootstrap.json": "fcffcb601c8e963c50a8a668b3f89880eee090e1d90f2219c779013699cc7fa3" + }, + "diagnosis": "Original uv 0.9.11 installer fetch failed before model execution: curl exit 35, SSL_ERROR_SYSCALL connecting to releases.astral.sh:443. Native NetworkConnectionError retained.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "task_ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "job_name": "dsh-dev34-pytorch-model-recovery-20261030", + "trial": "pytorch-model-recovery__VjqabNQ", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-recovery-20261030.json", + "native_reward": null, + "native_exception_type": "AgentSetupTimeoutError", + "standard_summary_status": "setup_failed", + "standard_summary_rewards": null, + "analytical_classification": "unscored_setup_error", + "analytical_score": null, + "boundary_status": "unconfirmed", + "process_supervision": {}, + "sdk_finish_reason": null, + "pytest": null, + "ctrf_summary": null, + "verifier_attempts": null, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:32:25.667665Z", + "finished_at": "2026-10-31T12:32:27.766588Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:32:27.766631Z", + "finished_at": "2026-10-31T13:02:27.001215Z" + }, + "agent_execution": null, + "verifier": null + }, + "agent_seconds": null, + "usage": { + "n_input_tokens": null, + "n_output_tokens": null, + "n_cache_tokens": null, + "cost_usd": null, + "raw_deepseek_metadata": null + }, + "operational_metadata": { + "event_types": {}, + "event_count": 0, + "tool_calls": 0, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 178.33929 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-pytorch-model-recovery-20261030.json": "874a0579bfd137c5c5ee6cf6741598cc629465e16567d957745b0a622ca7dc07", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/result.json": "a46c3c8544003a4b9b0a5e7166d38a3c511027b9f47f28e1e698f5ce3b3097ec", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/manifest.json": "dd319d8b72007f38327168b237035984477c465e9c685a05b8773d40ea67bb38", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030-input/config.json": "0b08782fc4919bf750aee2e8df7f2fec1ff83259d238a5cd7d4f4758a05d4613", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/trial.log": "649070006187e43bed47263196e7e41a2418e42b46968a5321147b7b9d9c0ced", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/exception.txt": "06df7743d111e65239e3a57cf415f3c566fd3a4ef2c91bf643d787c75505105f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-pytorch-model-recovery-20261030/pytorch-model-recovery__VjqabNQ/agent/bootstrap.json": "569634e2774a66f6a021d30250a11667067e2fb7503586e9cffd6f8eadacdb75" + }, + "diagnosis": "The formal job exceeded the 1800-second setup limit during verifier_dependencies. SDK initialization and supervision probe passed, but the registered dependency profile did not finish. No model execution occurred. The earlier installation-only profile check had passed; native AgentSetupTimeoutError retained.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/qemu-alpine-ssh", + "task_ref": "sha256:60b7050b0e0aa51641208cf65766743d340e59575db2c4d2f8628240846c2a28", + "job_name": "dsh-dev34-qemu-alpine-ssh-20261030", + "trial": "qemu-alpine-ssh__zFmhcaK", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-qemu-alpine-ssh-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 1, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 1, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793450287.6013296, + "stop": 1793450287.907775 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 11.137235495999448 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T12:32:24.916710Z", + "finished_at": "2026-10-31T12:32:27.016344Z" + }, + "agent_setup": { + "started_at": "2026-10-31T12:32:27.016384Z", + "finished_at": "2026-10-31T12:33:25.587144Z" + }, + "agent_execution": { + "started_at": "2026-10-31T12:33:25.587269Z", + "finished_at": "2026-10-31T12:37:55.802458Z" + }, + "verifier": { + "started_at": "2026-10-31T12:37:56.028301Z", + "finished_at": "2026-10-31T12:38:07.973295Z" + } + }, + "agent_seconds": 270.215189, + "usage": { + "n_input_tokens": 311197, + "n_output_tokens": 9504, + "n_cache_tokens": 298496, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 33, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 33, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 33, + "assistant/message": 33, + "tool/call": 32, + "tool/result": 32, + "step/end": 33, + "turn/end": 1 + }, + "event_count": 206, + "tool_calls": 32, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 654.860178 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-qemu-alpine-ssh-20261030.json": "9287c434ae61bea60ded0aee594ece8a251a5bb8be69db89497b4f21b926d9ca", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/result.json": "276a605348ba326675560d046b7d0e5d42596c8c2d58c9ae19b8bae966a8a2b8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/manifest.json": "d296f48ab339d781339af5fbe6fa9250b496b7d1c0d9636876fb27924cc276cf", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030-input/config.json": "066d9589181e89c1b553064064175377aa73d32f52613bbba680aa2b02695a11", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/agent/dsh-events.jsonl": "3d74bc6eb5928b841158625a3ea8f248c5203d7b8ebacb92e7535e8da383910f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/test-stdout.txt": "fa6dcd755e95aed0c733dcc234986140d131192411d8785160fafe7288887798", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-qemu-alpine-ssh-20261030/qemu-alpine-ssh__zFmhcaK/verifier/retry-summary.json": "cf571da57a07ad44dfbb678b31b038e53b9b1709234eb0cb5fc529bcb85c0d18" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/query-optimize", + "task_ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "job_name": "dsh-dev34-query-optimize-20261030", + "trial": "query-optimize__E5dWNng", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-query-optimize-20261030.json", + "native_reward": 0.0, + "native_exception_type": "NonZeroAgentExitCodeError", + "standard_summary_status": "execution_error", + "standard_summary_rewards": null, + "analytical_classification": "unscored_execution_error", + "analytical_score": null, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 1 + }, + "sdk_finish_reason": "error", + "pytest": { + "passed": 2, + "failed": 4, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 2, + "failed": 4, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793452129.5017295, + "stop": 1793452263.2794936 + }, + "verifier_attempts": { + "timeout_sec": 1800.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 214.38096272700022 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:03:08.437439Z", + "finished_at": "2026-10-31T13:03:10.567820Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:03:10.567862Z", + "finished_at": "2026-10-31T13:03:36.919967Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:03:36.920181Z", + "finished_at": "2026-10-31T13:07:28.574803Z" + }, + "verifier": { + "started_at": "2026-10-31T13:07:28.806280Z", + "finished_at": "2026-10-31T13:11:03.381861Z" + } + }, + "agent_seconds": 231.654622, + "usage": { + "n_input_tokens": 7681, + "n_output_tokens": 589, + "n_cache_tokens": 4096, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 3, + "missing_usage_records": 6, + "malformed_event_records": 0, + "usage_is_partial": true, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 4, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 3, + "assistant/message": 3, + "tool/call": 7, + "tool/result": 7, + "step/end": 4, + "assistant/attempt": 6, + "llm/retry": 5, + "llm/retry-started": 5, + "turn/end": 1 + }, + "event_count": 54, + "tool_calls": 7, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 317.974853 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-query-optimize-20261030.json": "b2e9089de82d0c6f52b5f345c6de388955b8852a37d809162a9a175aa8c25015", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/result.json": "214469bb2e966535105bb1478124f93dbb16186218ae7fa61feb3ad7a2289590", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/manifest.json": "51ff0c173afe9ee66c5b7d0dcfa0319a0f5703b302f3604bbd091714e80f1f3b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030-input/config.json": "2f89479bcaf4717dc24286cf5b6458c5beec5c5f9f23843f4770080ee8dc50ab", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/process-supervision.json": "eb8deb4559f5e566c55691ae383fa0249d537488bc446e7029e76b653396c566", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh-events.jsonl": "7e25269008a09aec00c241197a92ccc2b282732e350fe6c99f00d702c7f5850d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/test-stdout.txt": "61192bc4a8d7361a021abac1cea705206e9db281c3970f0a8057009bd65a66dd", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/verifier/retry-summary.json": "81ba64206851520023c8fa09a84a9d520e33f3fddbff0d49029bbe411f093db2", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/trial.log": "da8bcaf1adbe51c981a0d61f53f43d04addbb70740f25dc176f1ca6a9e1251e5", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/exception.txt": "fad5e5942af2769ba944129d9ebe76945b913d2e4996367ec211ee1050096882", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/dsh-result.json": "d9e17d8e9f5c60d0e263c11eec663a633550ece371f705b8f2b3f91148f1677f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-query-optimize-20261030/query-optimize__E5dWNng/agent/bootstrap.json": "89006797609af114ee0903b557c5ff434b3ef6071d8346100829e653fceb38ca" + }, + "diagnosis": "SDK terminated with finish_reason=error and a TRANSPORT error after six attempts of one provider request (five native request retries). Native NonZeroAgentExitCodeError retained. Official tests still ran (2 passed, 4 failed, missing SQL output), but the API-interrupted attempt is analytically unscored. No whole-trial retry.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/regex-log", + "task_ref": "sha256:802c16cfd132e6c457529cb864be5a757c1b23b6cadc57f2d01983cb0110292a", + "job_name": "dsh-dev34-regex-log-20261030", + "trial": "regex-log__RrrCtHf", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-log-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 1, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 1, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793451951.6564467, + "stop": 1793451951.6925104 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 43.60943662799946 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:03:07.470285Z", + "finished_at": "2026-10-31T13:03:09.692637Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:03:09.692697Z", + "finished_at": "2026-10-31T13:03:37.425613Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:03:37.425739Z", + "finished_at": "2026-10-31T13:05:07.745774Z" + }, + "verifier": { + "started_at": "2026-10-31T13:05:07.981430Z", + "finished_at": "2026-10-31T13:05:51.793464Z" + } + }, + "agent_seconds": 90.320035, + "usage": { + "n_input_tokens": 48284, + "n_output_tokens": 13394, + "n_cache_tokens": 45056, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 5, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 5, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 5, + "assistant/message": 5, + "tool/call": 4, + "tool/result": 4, + "step/end": 5, + "turn/end": 1 + }, + "event_count": 38, + "tool_calls": 4, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 157.046407 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-log-20261030.json": "efb5503decd193c0b0edc45142e68b6182fadf8662f3a197ee23ad928681b82a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/result.json": "7429bf2afe6a63ee8d962706b34c2d83dd397193d5d51e553d94ec8192b638c3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/manifest.json": "ad5558dac4408bcd388675a83a39005ccb88a09fac8eef3ae65423084b74162b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030-input/config.json": "530a7151f87605aff7f395a20357b1286868524a91bac5a0ae55c89410640ad5", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/agent/dsh-events.jsonl": "04211edfaba0b560f6731376296ca7b6b561005f22bb0af58b0ccf9f20b7261c", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/test-stdout.txt": "b75cc5ad308b4c261c27f0b693dd97a8d6a862af66036b5785a28b87f354a2ec", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-log-20261030/regex-log__RrrCtHf/verifier/retry-summary.json": "2bdc5193ca6685119434dcaa7060b08745575e52d92763a9f67052a83a35cc9e" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "task_ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "job_name": "dsh-dev34-torch-pipeline-parallelism-20261030", + "trial": "torch-pipeline-parallelism__zTzvuKq", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-pipeline-parallelism-20261030.json", + "native_reward": 0.0, + "native_exception_type": "AgentTimeoutError", + "standard_summary_status": "execution_error", + "standard_summary_rewards": null, + "analytical_classification": "unscored_verifier_error", + "analytical_score": null, + "boundary_status": "confirmed", + "process_supervision": { + "status": "cancelled", + "processes_stopped": true, + "stopped_processes": 8 + }, + "sdk_finish_reason": null, + "pytest": null, + "ctrf_summary": null, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 384.8488658469996 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:12:05.510333Z", + "finished_at": "2026-10-31T13:12:07.626414Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:12:07.626456Z", + "finished_at": "2026-10-31T13:12:36.149669Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:12:36.149860Z", + "finished_at": "2026-10-31T13:27:35.454731Z" + }, + "verifier": { + "started_at": "2026-10-31T13:27:35.690609Z", + "finished_at": "2026-10-31T13:33:59.103316Z" + } + }, + "agent_seconds": 899.304871, + "usage": { + "n_input_tokens": 545691, + "n_output_tokens": 34325, + "n_cache_tokens": 536960, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 23, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": true, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 23, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 23, + "assistant/message": 23, + "tool/call": 23, + "tool/result": 22, + "step/end": 22 + }, + "event_count": 145, + "tool_calls": 23, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": true, + "agent_seconds": 562.196283 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-pipeline-parallelism-20261030.json": "60fb09e7a9e852149551c7ec2930f4288785c6b4dc5451d486dcfc509b0c0d37", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/result.json": "bbd946c59c4aea9ed60f6233def0fa4d97ce45b77650b12798fa01ceaeb6ea56", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/manifest.json": "88a5be96e11de4d421b3751d836776344b26ba21e8297832fd3369b1f6debd9d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030-input/config.json": "0cda3c9547114934803d112b09228b6f415ecc73cb28d02a4668149a24536b46", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/process-supervision.json": "234c3369ed928b4691b85a2261e195dd42f2fcf5556a6cb292ace22db8b2c109", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/dsh-events.jsonl": "4ad4aca4726fb1a60989217de15d5c3ddc50de7fddcd0ba2f781e202a9d4c306", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/test-stdout.txt": "d92c853dd2b021a6c77661b03c422071d2bb1bffcd4db8fd5a0a44e3c20cc5ce", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/verifier/retry-summary.json": "c639c755a9d5d93f321985096563a85955f8293dc98a428fe59a0bc8e8f1231f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/trial.log": "7d5613e1d7303b60d682642d18c6cdde1228da9f4fd71c2ab1d6541fa182a88b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/exception.txt": "3db31c2d14b585a222dfe95002d19c8d715c4248f17b41224a6e95a9acc936a8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-pipeline-parallelism-20261030/torch-pipeline-parallelism__zTzvuKq/agent/bootstrap.json": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + }, + "diagnosis": "Native 900-second agent timeout was confirmed with 8 stopped processes. Official verifier then failed downloading nvidia-cusolver-cu12==11.7.1.2 required by torch==2.7.0, before pytest began. Dependency error makes the raw zero analytically null; native timeout retained.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "task_ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "job_name": "dsh-dev34-torch-tensor-parallelism-20261030", + "trial": "torch-tensor-parallelism__ZVRUdrm", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-tensor-parallelism-20261030.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "unscored_verifier_error", + "analytical_score": null, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": null, + "ctrf_summary": null, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "timeout", + "duration_seconds": 900.0019550929992 + }, + { + "attempt": 2, + "status": "completed", + "return_code": 0, + "duration_seconds": 432.4585692649998 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:12:04.063481Z", + "finished_at": "2026-10-31T13:12:06.181914Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:12:06.181972Z", + "finished_at": "2026-10-31T13:12:40.496141Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:12:40.496349Z", + "finished_at": "2026-10-31T13:14:51.775456Z" + }, + "verifier": { + "started_at": "2026-10-31T13:14:52.006261Z", + "finished_at": "2026-10-31T13:37:02.586882Z" + } + }, + "agent_seconds": 131.279107, + "usage": { + "n_input_tokens": 87318, + "n_output_tokens": 17531, + "n_cache_tokens": 83456, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 8, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 8, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 8, + "assistant/message": 8, + "tool/call": 7, + "tool/result": 7, + "step/end": 8, + "turn/end": 1 + }, + "event_count": 56, + "tool_calls": 7, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 90.783629 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-torch-tensor-parallelism-20261030.json": "ce23dc2c2eae0656f0a93120f2cf3185d383753534d7dece3657dec23bf68ff0", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/result.json": "d06d9acad5f9df54a7430dedd63542e88f240dedebf0517b19980adb22416e39", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/manifest.json": "b073682154393d82f6b6de279ac58f2fbb4eb9e39f9ff6967bd5b3fd89980e3c", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030-input/config.json": "7b136d0a86d5cb62dd83805a320a76f288880f27d927f81ab0cfeffe6c2b4a6a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh-events.jsonl": "4619ce1d5a34be369ba4aa4b3d9bbe54472096aee99dfcdadcb53c92f5f4d6da", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/test-stdout.txt": "001335f99efb1ac4511114fd5274d30802fbf47e78a69922f8700dcdcd742c8e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/verifier/retry-summary.json": "c93db7d2a7b90856b956a7c8704b7f02b380503c5a3b5feca67b0e81741045f6", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/trial.log": "ae88953072e2d4ea99824bb29d4e74512900ecd25bb82982e50fc8515d3a6bcf", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/dsh-result.json": "082539d153b799604aadd41eedd3d4686cf5af202f5e86170214f860b372c8d6", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-torch-tensor-parallelism-20261030/torch-tensor-parallelism__ZVRUdrm/agent/bootstrap.json": "36699c5279f25d50f5b87aa7cc65a98e6c3ee72ff24f5ec4a88bc3751509d809" + }, + "diagnosis": "First verifier attempt hit the native 900-second timeout. The second attempt failed downloading torch==2.7.0 with UV_HTTP_TIMEOUT=30 before pytest began. Raw zero retained; analytical score null.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/write-compressor", + "task_ref": "sha256:d9ddd9a8e925e2c566b37b2492cbf995afecefe58874e4043ef78d7f3c892c7e", + "job_name": "dsh-dev34-write-compressor-20261030", + "trial": "write-compressor__YhbFGda", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-write-compressor-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 3, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793454080.788398, + "stop": 1793454080.9124975 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 17.64990917599971 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:38:02.078384Z", + "finished_at": "2026-10-31T13:38:04.223488Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:38:04.223529Z", + "finished_at": "2026-10-31T13:38:39.263025Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:38:39.263145Z", + "finished_at": "2026-10-31T13:41:02.888407Z" + }, + "verifier": { + "started_at": "2026-10-31T13:41:03.125406Z", + "finished_at": "2026-10-31T13:41:20.968583Z" + } + }, + "agent_seconds": 143.625262, + "usage": { + "n_input_tokens": 113738, + "n_output_tokens": 17658, + "n_cache_tokens": 110080, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 11, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 11, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 11, + "assistant/message": 11, + "tool/call": 10, + "tool/result": 10, + "step/end": 11, + "turn/end": 1 + }, + "event_count": 74, + "tool_calls": 10, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 304.111666 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-write-compressor-20261030.json": "a0856e05c54fc115e26f9272f30fabfd821cc9848c09b49dff5f7d319da542e4", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/result.json": "9282ab7ec0f5b980db603031a78dddd20add13400918c58d720e8d05f68f1222", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/manifest.json": "1422633177e16f08b3176abac96fb3c40ca252f48668a6b1780b9d39583a1c68", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030-input/config.json": "5bffbeeea5a40725c5bf95c87093eaadcc52b59e9d555472a55e95ab52a17d66", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/agent/dsh-events.jsonl": "70014d4db3875f0d0a369c10e47aa09d7654c21de1e4f5be407c8f118227bbd9", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/test-stdout.txt": "166f15fa270868102a648fd0a846285db328a9c91964451a36c4fed8937c89ce", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-write-compressor-20261030/write-compressor__YhbFGda/verifier/retry-summary.json": "1617f955d7afcd632659daf0e6109b63b73ab0689e175fa10e9ffec7bde45e8c" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/constraints-scheduling", + "task_ref": "sha256:91dd6f87e1ee508328d5aafbd99a6bd1ccc47bc22671b4aaa4060bda492deb53", + "job_name": "dsh-dev34-constraints-scheduling-20261030", + "trial": "constraints-scheduling__SZ9L9Ee", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-constraints-scheduling-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 3, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793453990.8383439, + "stop": 1793453991.2194958 + }, + "verifier_attempts": { + "timeout_sec": 1200.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 36.992148000999805 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:38:03.447083Z", + "finished_at": "2026-10-31T13:38:05.531232Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:38:05.531290Z", + "finished_at": "2026-10-31T13:38:49.569717Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:38:49.569843Z", + "finished_at": "2026-10-31T13:39:13.897804Z" + }, + "verifier": { + "started_at": "2026-10-31T13:39:14.125439Z", + "finished_at": "2026-10-31T13:39:51.312049Z" + } + }, + "agent_seconds": 24.327961, + "usage": { + "n_input_tokens": 9461, + "n_output_tokens": 2287, + "n_cache_tokens": 6656, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 3, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 3, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 3, + "assistant/message": 3, + "tool/call": 2, + "tool/result": 2, + "step/end": 3, + "turn/end": 1 + }, + "event_count": 26, + "tool_calls": 2, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 22.995093 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-constraints-scheduling-20261030.json": "f3427b83f8e867c11ec00983493c92180a837ce6746c85b454a52841c0bf83ba", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/result.json": "6790ec0f4a1f9ccb9b191af8361ae8d77bf7f59794aa2f4e04fc0238d85043c9", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/manifest.json": "49b36ec4775750b7cd22b72de420cceaead5c20a53502a75bd3f9976049bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030-input/config.json": "79c6b95efc3d932b92f3d4f4054faa08cc1407b6d79515f16f56844724d85c5b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/agent/dsh-events.jsonl": "71fdd2595bb7d4af395122260ddbc6dacfdf730d8f35184a351f098237cd879b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/test-stdout.txt": "cc98ac573e8853538edc9451bbb2c12830300030b925ab658cc2c1c3b64cebf3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-constraints-scheduling-20261030/constraints-scheduling__SZ9L9Ee/verifier/retry-summary.json": "f58a53fcb7b5c1ad7d8666b587d047bc7c01fcdcadf0f3a738f05e7187345623" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/custom-memory-heap-crash", + "task_ref": "sha256:61e021e0304c9818a9be2c9173c8a0938ce5d136e95c5422e099f1fdb2dc080e", + "job_name": "dsh-dev34-custom-memory-heap-crash-20261030", + "trial": "custom-memory-heap-crash__eHwske6", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-custom-memory-heap-crash-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 6, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 6, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793454339.2450511, + "stop": 1793454342.8759143 + }, + "verifier_attempts": { + "timeout_sec": 1800.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 22.11033539300115 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:42:40.777536Z", + "finished_at": "2026-10-31T13:42:42.867103Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:42:42.867167Z", + "finished_at": "2026-10-31T13:43:09.074253Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:43:09.074393Z", + "finished_at": "2026-10-31T13:45:20.823967Z" + }, + "verifier": { + "started_at": "2026-10-31T13:45:21.086192Z", + "finished_at": "2026-10-31T13:45:42.932227Z" + } + }, + "agent_seconds": 131.749574, + "usage": { + "n_input_tokens": 414021, + "n_output_tokens": 12496, + "n_cache_tokens": 386432, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 21, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 21, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 21, + "assistant/message": 21, + "tool/call": 20, + "tool/result": 20, + "step/end": 21, + "turn/end": 1 + }, + "event_count": 134, + "tool_calls": 20, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 96.213971 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-custom-memory-heap-crash-20261030.json": "91243e3c6af247f91c7dbfb0bb710cca80e0e0f3055c0fc509800e879236ca40", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/result.json": "50c651e46529ccb383769ca6a63e5d36d37bf0ef57fd5d412eabd623cb3f0e40", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/manifest.json": "e5f5ba1a234be82c7021354b6052a1717451e0de1e84e57e668823ecf88bdea9", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030-input/config.json": "30999de9f48bb348a50ab8c27e23185ec6c1d6e20d39c3d830ed9d445d5a454b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/agent/dsh-events.jsonl": "38807afa25b865ea21d58c186b5dc66cac08cbc4881cebb9ae0a12196829162e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/test-stdout.txt": "40dc429d0e27f09c31a68177111c6fe452d135367444fc487d8f7921d1a9e632", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-custom-memory-heap-crash-20261030/custom-memory-heap-crash__eHwske6/verifier/retry-summary.json": "809e75cacbffd614d0e49daefbc4ccf3525152b7c75a710c9d5027745400f0f8" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/dna-assembly", + "task_ref": "sha256:e41a8e94d86019949b08d3b5f88a85f6d943ba0fd85d5e1d5ebb95cb8f66223f", + "job_name": "dsh-dev34-dna-assembly-20261030", + "trial": "dna-assembly__GNnnHxn", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-dna-assembly-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 1, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 1, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793455159.2723281, + "stop": 1793455159.317828 + }, + "verifier_attempts": { + "timeout_sec": 1800.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 16.296766847999606 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T13:42:40.711002Z", + "finished_at": "2026-10-31T13:42:42.814097Z" + }, + "agent_setup": { + "started_at": "2026-10-31T13:42:42.814158Z", + "finished_at": "2026-10-31T13:43:18.205674Z" + }, + "agent_execution": { + "started_at": "2026-10-31T13:43:18.205875Z", + "finished_at": "2026-10-31T13:59:02.683930Z" + }, + "verifier": { + "started_at": "2026-10-31T13:59:02.915894Z", + "finished_at": "2026-10-31T13:59:19.416894Z" + } + }, + "agent_seconds": 944.478055, + "usage": { + "n_input_tokens": 4105050, + "n_output_tokens": 83418, + "n_cache_tokens": 4036352, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 56, + "missing_usage_records": 1, + "malformed_event_records": 0, + "usage_is_partial": true, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 56, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 56, + "assistant/message": 56, + "tool/call": 55, + "tool/result": 55, + "step/end": 56, + "assistant/attempt": 1, + "llm/retry": 1, + "llm/retry-started": 1, + "turn/end": 1 + }, + "event_count": 347, + "tool_calls": 55, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 763.268852 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-dna-assembly-20261030.json": "bce94de2fc9b3566fbf30ddcb424e5a5a95d68c124cf7eea11cafc14afe1f818", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/result.json": "6594d6b20c239a591ea5a63934e301bdb5faa48d136a9e670540bcb984527e8a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/manifest.json": "446398349da934e87a61312afab8f6acfcc311fa8a26264de0eb964c0be28443", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030-input/config.json": "af06e540459232eded056d2429ff9bea502f8b94ff9667fdf144b93b6bcf8407", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/agent/dsh-events.jsonl": "fc076f25f57458a5e5bcd86ec2e38590e862e520edb9aa3333701351c1f6c67b", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/test-stdout.txt": "344bf45577a2c7715621a2f3e7a669dce177019580c4a2a70c723f7cda4682ce", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-dna-assembly-20261030/dna-assembly__GNnnHxn/verifier/retry-summary.json": "62e2ae1d81da0eeaf24315c1b8c5de1a47b0d3dc17e0fd5cb84fff6e75fa10aa" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/llm-inference-batching-scheduler", + "task_ref": "sha256:a3bf47589118daec5124fe3689b4439802c805b2f4ed9d402a48ae73ca2fea67", + "job_name": "dsh-dev34-llm-inference-batching-scheduler-20261030", + "trial": "llm-inference-batching-scheduler__u6RXWQU", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-llm-inference-batching-scheduler-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 6, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 6, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793455789.7117288, + "stop": 1793455789.7930355 + }, + "verifier_attempts": { + "timeout_sec": 1800.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 34.83602994800094 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T14:00:08.828668Z", + "finished_at": "2026-10-31T14:00:10.992907Z" + }, + "agent_setup": { + "started_at": "2026-10-31T14:00:10.992950Z", + "finished_at": "2026-10-31T14:00:50.842390Z" + }, + "agent_execution": { + "started_at": "2026-10-31T14:00:50.842586Z", + "finished_at": "2026-10-31T14:09:14.607622Z" + }, + "verifier": { + "started_at": "2026-10-31T14:09:14.839318Z", + "finished_at": "2026-10-31T14:09:49.870661Z" + } + }, + "agent_seconds": 503.765036, + "usage": { + "n_input_tokens": 1049957, + "n_output_tokens": 35106, + "n_cache_tokens": 1012480, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 28, + "missing_usage_records": 1, + "malformed_event_records": 0, + "usage_is_partial": true, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 28, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 28, + "assistant/message": 28, + "tool/call": 30, + "tool/result": 30, + "step/end": 28, + "assistant/attempt": 1, + "llm/retry": 1, + "llm/retry-started": 1, + "turn/end": 1 + }, + "event_count": 185, + "tool_calls": 30, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 337.746791 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-llm-inference-batching-scheduler-20261030.json": "dcec6c590dcc55800de1ef96a5ac0565a941be4bfcaae4b5996392a832851295", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/result.json": "6e1542ae91cf1eba19ff628595fdb4922bb487ab10dec811c9af61f8c8e0a10e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/manifest.json": "d1bfa964cc85d0f4112bc2403a619f4beee2da7f861c9c1fcbcea8c2a2c36086", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030-input/config.json": "b0b250afcac986756f36b26dd7e4c82af7cd8d5d827ce1e2fada7b573e280aef", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/agent/dsh-events.jsonl": "cfe9604b53d98704bdf0088e29c85925a4c3bfaf545a0e229ff7eecc6b0d2586", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/test-stdout.txt": "c1b8385af6f24121316013915f21817b1b0bd8dcee6b8623ce60e217ea5e880f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-llm-inference-batching-scheduler-20261030/llm-inference-batching-scheduler__u6RXWQU/verifier/retry-summary.json": "4e304dd6218d48212f724c93b1b2c5fc1aff4aca1f64ef748aab6d8782397aa3" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/path-tracing", + "task_ref": "sha256:cf56094c881a488b27e9f204a638a7e78ed7d55e12dc3064108c93357190314c", + "job_name": "dsh-dev34-path-tracing-20261030", + "trial": "path-tracing__NacuN7E", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-path-tracing-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 5, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 5, + "passed": 5, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793455545.9323423, + "stop": 1793455553.9861736 + }, + "verifier_attempts": { + "timeout_sec": 1800.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 85.78095239700087 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T14:00:09.124988Z", + "finished_at": "2026-10-31T14:00:11.166122Z" + }, + "agent_setup": { + "started_at": "2026-10-31T14:00:11.166165Z", + "finished_at": "2026-10-31T14:00:38.072677Z" + }, + "agent_execution": { + "started_at": "2026-10-31T14:00:38.072809Z", + "finished_at": "2026-10-31T14:04:27.884831Z" + }, + "verifier": { + "started_at": "2026-10-31T14:04:28.110727Z", + "finished_at": "2026-10-31T14:05:54.082967Z" + } + }, + "agent_seconds": 229.812022, + "usage": { + "n_input_tokens": 806310, + "n_output_tokens": 20991, + "n_cache_tokens": 761344, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 28, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 28, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 28, + "assistant/message": 28, + "tool/call": 27, + "tool/result": 27, + "step/end": 28, + "turn/end": 1 + }, + "event_count": 176, + "tool_calls": 27, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 133.05033 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-path-tracing-20261030.json": "1e99fa743f588514d4d15d397341247f118eae2c5e2f30faf8bd1781d3a64e31", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/result.json": "00ddde593f3721a51d4629de0d322dc56127daafb25ed0b6a9c382b64c4e1a85", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/manifest.json": "b854a834dec4f93014e92aab9fb9488c894d463064e9365efb860c6d7401215e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030-input/config.json": "269089d8ad7d35f433b7a69a64e20cf34685301ad1a6958718344b34536f21d5", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/agent/dsh-events.jsonl": "8faa97f012c83f1d8770c433b82fd50d068d045e89c7de1a433ce6ab8c6f023d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/test-stdout.txt": "34b5428a6ec54e07c1f7b186fa761f233e5eb6bc79d630fe81aa2b424c6c0617", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-path-tracing-20261030/path-tracing__NacuN7E/verifier/retry-summary.json": "007fc2888cc95d1ff358de21099c658aee73c862a5d3ffb78b0fd8e7a94b8007" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/schemelike-metacircular-eval", + "task_ref": "sha256:58130c2166c3115276dc8592f358e326ff2d81ea852e3d88636c82fd1dff57e6", + "job_name": "dsh-dev34-schemelike-metacircular-eval-20261030", + "trial": "schemelike-metacircular-eval__V7oLYdt", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-schemelike-metacircular-eval-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 1, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 1, + "passed": 1, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793456806.7036533, + "stop": 1793456876.4352946 + }, + "verifier_attempts": { + "timeout_sec": 2400.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 89.38283998400038 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T14:10:44.591253Z", + "finished_at": "2026-10-31T14:10:46.748955Z" + }, + "agent_setup": { + "started_at": "2026-10-31T14:10:46.748999Z", + "finished_at": "2026-10-31T14:11:23.908946Z" + }, + "agent_execution": { + "started_at": "2026-10-31T14:11:23.909081Z", + "finished_at": "2026-10-31T14:26:26.692402Z" + }, + "verifier": { + "started_at": "2026-10-31T14:26:26.924265Z", + "finished_at": "2026-10-31T14:27:56.507433Z" + } + }, + "agent_seconds": 902.783321, + "usage": { + "n_input_tokens": 1350725, + "n_output_tokens": 40156, + "n_cache_tokens": 1326080, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 37, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 37, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 37, + "assistant/message": 37, + "tool/call": 36, + "tool/result": 36, + "step/end": 37, + "turn/end": 1 + }, + "event_count": 230, + "tool_calls": 36, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 800.124038 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-schemelike-metacircular-eval-20261030.json": "91a8c4c0b5611c775b5c122414695efcf5c722d96dbf5afc21b2ed91dd050734", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/result.json": "ce6dcb948972f30008bc84bf3f25bc86fac08f0a4d369210ee89327a7f9339c9", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/manifest.json": "d6111a6bb4ec04f97228747719efea70040f30e6f05057bec6b142e791187eaf", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030-input/config.json": "55143fc6d17f7320ee39534c2480ef97e1d76cd9d3911763972399b9d516e76e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/agent/dsh-events.jsonl": "f57fe30e205fa70c52674163266c222ee11fd7718ca564abf0b1a563738df0af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/test-stdout.txt": "4efcd17d33883ef230af8cb5a3ee3c2d2eeabfb3f0e1982dc0a5ca19ce42955f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-schemelike-metacircular-eval-20261030/schemelike-metacircular-eval__V7oLYdt/verifier/retry-summary.json": "e41c3df547232ff74d3c6a9955506072a71d29bfdbaebbf6318abac534a13ccf" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/caffe-cifar-10", + "task_ref": "sha256:7b0045106d7d5af724efe96b610ba64f7893f5c88528401c573c4d47e384e2bf", + "job_name": "dsh-dev34-caffe-cifar-10-20261030", + "trial": "caffe-cifar-10__tmpc3nU", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-caffe-cifar-10-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 6, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 6, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793458135.328734, + "stop": 1793458174.6339345 + }, + "verifier_attempts": { + "timeout_sec": 1200.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 75.65802802100006 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T14:10:44.667235Z", + "finished_at": "2026-10-31T14:10:46.749506Z" + }, + "agent_setup": { + "started_at": "2026-10-31T14:10:46.749562Z", + "finished_at": "2026-10-31T14:11:38.096494Z" + }, + "agent_execution": { + "started_at": "2026-10-31T14:11:38.096624Z", + "finished_at": "2026-10-31T14:48:18.639452Z" + }, + "verifier": { + "started_at": "2026-10-31T14:48:18.871054Z", + "finished_at": "2026-10-31T14:49:34.727555Z" + } + }, + "agent_seconds": 2200.542828, + "usage": { + "n_input_tokens": 1864152, + "n_output_tokens": 15478, + "n_cache_tokens": 1807872, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 55, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 55, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 55, + "assistant/message": 55, + "tool/call": 55, + "tool/result": 55, + "step/end": 55, + "turn/end": 1 + }, + "event_count": 340, + "tool_calls": 55, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed_after_confirmed_agent_timeout", + "reference_exposure": false, + "agent_seconds": 3600.412731 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-caffe-cifar-10-20261030.json": "3e5c92478fc5b7cc715f3f9f10147d67d8b1d55c4e5210a56ad1949819c57612", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/result.json": "09d2038486503d272f70461f6d21f39f905ebb5b73bb2a35b2c36d7cbb8de5c7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/manifest.json": "e15b80c308deaea2c9d073f59599096b6379730a25029b0e6dc4b306300089a6", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030-input/config.json": "9f5463c2fe65e195928ed9eab0033d2f8fd0f6f807f98667f321f49a1c7eba10", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/agent/dsh-events.jsonl": "9f3d9966e0fbee53a604346d14d3e6e147795087c3cf652a95007531a51a18c7", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/test-stdout.txt": "e9157af437d576f9d81e98ee3314eb6e702832a842a645a62340305541a637fa", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-caffe-cifar-10-20261030/caffe-cifar-10__tmpc3nU/verifier/retry-summary.json": "ef3f73839e37b98d0d43a86d1e73e0075189596159ad1f8c534b87a60dc40a80" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/circuit-fibsqrt", + "task_ref": "sha256:9bcffe1054bb33249aa578a9a2a74f3c8cca66b0cb7aa1328233f1d31822aae3", + "job_name": "dsh-dev34-circuit-fibsqrt-20261030", + "trial": "circuit-fibsqrt__y4RQ7w4", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-circuit-fibsqrt-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 3, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793458740.5809975, + "stop": 1793458763.698463 + }, + "verifier_attempts": { + "timeout_sec": 3600.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 41.792968114999894 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T14:50:31.382480Z", + "finished_at": "2026-10-31T14:50:33.481811Z" + }, + "agent_setup": { + "started_at": "2026-10-31T14:50:33.481873Z", + "finished_at": "2026-10-31T14:51:00.397983Z" + }, + "agent_execution": { + "started_at": "2026-10-31T14:51:00.398105Z", + "finished_at": "2026-10-31T14:58:41.455646Z" + }, + "verifier": { + "started_at": "2026-10-31T14:58:41.698748Z", + "finished_at": "2026-10-31T14:59:23.773857Z" + } + }, + "agent_seconds": 461.057541, + "usage": { + "n_input_tokens": 783666, + "n_output_tokens": 54639, + "n_cache_tokens": 754560, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 18, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 18, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 18, + "assistant/message": 18, + "tool/call": 17, + "tool/result": 17, + "step/end": 18, + "turn/end": 1 + }, + "event_count": 116, + "tool_calls": 17, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 239.038995 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-circuit-fibsqrt-20261030.json": "cf7037c85fda1367bc8fc2270f839c3b6527d7af22ca0cf906b123181d06260d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/result.json": "0b575ea99418eed3108dd653f769b67d8bd58e8e7e9b790f76551114c72d01cc", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/manifest.json": "144d39b5e20bf5333fa90dbcb759eb8a8b78833460b856bde474236828cb8bc3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030-input/config.json": "05a46047d538cc18f7433d60ee9341b4dc871923402dbddeb2b514b3293398ad", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/agent/dsh-events.jsonl": "60982612ae2d7b8ad86c2ddbe04afb5486971bcb627b9d633860edc1eee91310", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/test-stdout.txt": "56a10f48113ce0398a1a03d05a3c2c68ecc8bc257732c9370a7d5a316b602a3e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-circuit-fibsqrt-20261030/circuit-fibsqrt__y4RQ7w4/verifier/retry-summary.json": "9165e04d7545e0e6357a7851c7d6454771791c957049b5f5fb7ef16257c7a13b" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/mteb-leaderboard", + "task_ref": "sha256:484f6d7008a05b5b8640fc6618a384b8c9447cd76f85416c8a595028d29bff9c", + "job_name": "dsh-dev34-mteb-leaderboard-20261030", + "trial": "mteb-leaderboard__hbrgVf4", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-mteb-leaderboard-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 2, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 2, + "passed": 2, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793459447.5703776, + "stop": 1793459447.605211 + }, + "verifier_attempts": { + "timeout_sec": 3600.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 18.402055131000452 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T14:50:30.437750Z", + "finished_at": "2026-10-31T14:50:32.614818Z" + }, + "agent_setup": { + "started_at": "2026-10-31T14:50:32.614877Z", + "finished_at": "2026-10-31T14:50:56.632654Z" + }, + "agent_execution": { + "started_at": "2026-10-31T14:50:56.632808Z", + "finished_at": "2026-10-31T15:10:28.865703Z" + }, + "verifier": { + "started_at": "2026-10-31T15:10:29.099602Z", + "finished_at": "2026-10-31T15:10:47.693184Z" + } + }, + "agent_seconds": 1172.232895, + "usage": { + "n_input_tokens": 4995195, + "n_output_tokens": 25006, + "n_cache_tokens": 4893824, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 83, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 83, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 83, + "assistant/message": 83, + "tool/call": 82, + "tool/result": 82, + "step/end": 83, + "turn/end": 1 + }, + "event_count": 506, + "tool_calls": 82, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 667.653236 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-mteb-leaderboard-20261030.json": "19eac1bdc69f1a74dbe0d0897e2bb1aeb7c312b50ee21299bdc9aa721290140e", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/result.json": "37676ebee576524bf8103b0bdc29c97c587f954d58122958ebe0983eb3b686a8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/manifest.json": "144d39b5e20bf5333fa90dbcb759eb8a8b78833460b856bde474236828cb8bc3", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030-input/config.json": "d9ec1602d13d24b36373d4f9655f136a4fd316e1ea3670b80a9ed881ffea4804", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/agent/dsh-events.jsonl": "de54ed6deb6e7029b2f805b488e76b83df79ac4d543a423dfb6840e665c42285", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/test-stdout.txt": "f128f0efc5ee5a556e06100cacd777393fbcc796c16bee8df99a01271715058a", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-mteb-leaderboard-20261030/mteb-leaderboard__hbrgVf4/verifier/retry-summary.json": "048f14f0da7b01d7574021ca3d4613441ad00683c40a2af09ea0c79225329356" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": true + } + }, + { + "task": "terminal-bench/regex-chess", + "task_ref": "sha256:e763e0ac1c9759081af0a4a82ba51b8cf9ae5485a93de3bbe42d7d344597bd78", + "job_name": "dsh-dev34-regex-chess-20261030", + "trial": "regex-chess__qF2PDED", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-chess-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 4, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 4, + "passed": 4, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793461459.0224414, + "stop": 1793461553.5213957 + }, + "verifier_attempts": { + "timeout_sec": 3600.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 117.8494001729996 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T15:11:31.311339Z", + "finished_at": "2026-10-31T15:11:33.533790Z" + }, + "agent_setup": { + "started_at": "2026-10-31T15:11:33.533911Z", + "finished_at": "2026-10-31T15:12:07.552131Z" + }, + "agent_execution": { + "started_at": "2026-10-31T15:12:07.552264Z", + "finished_at": "2026-10-31T15:43:55.287315Z" + }, + "verifier": { + "started_at": "2026-10-31T15:43:55.534631Z", + "finished_at": "2026-10-31T15:45:53.597189Z" + } + }, + "agent_seconds": 1907.735051, + "usage": { + "n_input_tokens": 4327394, + "n_output_tokens": 82437, + "n_cache_tokens": 4249728, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 61, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 61, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 61, + "assistant/message": 61, + "tool/call": 60, + "tool/result": 60, + "step/end": 61, + "turn/end": 1 + }, + "event_count": 374, + "tool_calls": 60, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 1618.237896 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-regex-chess-20261030.json": "64840e42fb8b79e179382b2d8d31d89b836a6856bb93c6d618ce12697fd3e487", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/result.json": "6e7a80996866db583f4b08991823055f2f0971dd26c7b942686c918a00260dae", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/manifest.json": "0f89d986e9251cc5815ed577f9e67de05c9b19136a9a54690f7798a6b6764426", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030-input/config.json": "96614fdb59d09909d7343c3966868c5663476f23f9319da85b3c59c596935ef9", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/agent/dsh-events.jsonl": "f96e82c2d006c2b6c5855adbf49691f3e4ee3b1f11a11b488762049d1835edb4", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/test-stdout.txt": "b4409c4e80cd3272c7d52ac37e04e70b8ac10040361baceeeb2665a3fcbb90eb", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-regex-chess-20261030/regex-chess__qF2PDED/verifier/retry-summary.json": "df85bc4277f0973448621ef8d45897f83614a5c430b2827f2bbd4a82f59dd92d" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + }, + { + "task": "terminal-bench/build-pov-ray", + "task_ref": "sha256:4f08dab602fe2e9e2f42051cc6aef684a88cf2822e8b499b3ca19e670759fddb", + "job_name": "dsh-dev34-build-pov-ray-20261030", + "trial": "build-pov-ray__D5bxRpk", + "arm": "deepseek", + "source_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "native_root": "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk", + "record": "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-pov-ray-20261030.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 3, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793460526.2365377, + "stop": 1793460532.6995895 + }, + "verifier_attempts": { + "timeout_sec": 12000.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 118.68512292300147 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-10-31T15:11:31.562057Z", + "finished_at": "2026-10-31T15:11:33.677264Z" + }, + "agent_setup": { + "started_at": "2026-10-31T15:11:33.677303Z", + "finished_at": "2026-10-31T15:12:01.753605Z" + }, + "agent_execution": { + "started_at": "2026-10-31T15:12:01.753828Z", + "finished_at": "2026-10-31T15:26:53.691585Z" + }, + "verifier": { + "started_at": "2026-10-31T15:26:53.918845Z", + "finished_at": "2026-10-31T15:28:52.814715Z" + } + }, + "agent_seconds": 891.937757, + "usage": { + "n_input_tokens": 698524, + "n_output_tokens": 7543, + "n_cache_tokens": 655872, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 33, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 33, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 33, + "assistant/message": 33, + "tool/call": 32, + "tool/result": 32, + "step/end": 33, + "turn/end": 1 + }, + "event_count": 206, + "tool_calls": 32, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 267.416658 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev34-20261030-orchestration/dsh-dev34-build-pov-ray-20261030.json": "d36751765bc73718db69e4aad091f44c690c38a045d437028867b434865508bf", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/result.json": "1a55e19d4d065a03cf0e8463a23fcfb2f01290117d32a75d2d740dfb51853d5d", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/manifest.json": "d67acc20f268c504f07405fe5a4fb004985fe5c881127e6ee94147feebfe3006", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030-input/config.json": "1b0f9b61f0e331332ede265d919619df18bd37dcf6f53280ab5c8ed0d570129f", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/agent/dsh-events.jsonl": "aa213998b0adb6b57a19d2b73a70fd00f9070c201b82dc99a4cb600276efdfe8", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/test-stdout.txt": "e0db42279d971b35469d51ac5e20f9c3de9ff9143219300819e47b220ec829ad", + "jobs/deepseek-harness-runtime-source/jobs/dsh-dev34-build-pov-ray-20261030/build-pov-ray__D5bxRpk/verifier/retry-summary.json": "c22fef9936b4b5e2f7c3160ab737580603c2c5758819750ea79f7453d11d7bab" + }, + "diagnosis": "Original official verifier completed and awarded reward one.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "upstream_application_material": false + } + } + ] +} diff --git a/benchmarks/harbor/scripts/probe_deepseek_lifecycle.py b/benchmarks/harbor/scripts/probe_deepseek_lifecycle.py new file mode 100644 index 0000000..9efff5b --- /dev/null +++ b/benchmarks/harbor/scripts/probe_deepseek_lifecycle.py @@ -0,0 +1,80 @@ +"""Probe sdk-minimal background-server lifetime with upstream keyless fixtures.""" +import argparse +from datetime import datetime, timezone +import hashlib +import importlib.util +import json +from pathlib import Path +import socket +import sys +import tempfile +import time +import urllib.error +import urllib.request + + +def main(): + parser = argparse.ArgumentParser(description=__doc__) + parser.add_argument('--upstream-source', type=Path, required=True) + parser.add_argument('--runtime', type=Path, required=True) + args = parser.parse_args() + source, runtime = args.upstream_source.resolve(), args.runtime.resolve() + sys.path.insert(0, str(source / 'python/sdk/src')) + from deepseek_harness import DeepSeekHarness + fixture = source / 'scripts/smoke-python-runtime.py' + spec = importlib.util.spec_from_file_location('upstream_smoke', fixture) + smoke = importlib.util.module_from_spec(spec) + spec.loader.exec_module(smoke) + with socket.socket() as listener: + listener.bind(('127.0.0.1', 0)) + port = listener.getsockname()[1] + calls = 0 + def completion(_body): + nonlocal calls + calls += 1 + if calls == 1: + return smoke.tool_call_chunks('background-server', 'bash', { + 'command': f'nohup python3 -m http.server {port} --bind 127.0.0.1 > server.log 2>&1 & echo $! > server.pid; sleep 1', + }) + return smoke.text_chunks('Lifecycle probe completed.') + smoke.completion_chunks = completion + def reachable(): + try: + with urllib.request.urlopen(f'http://127.0.0.1:{port}/', timeout=1) as response: + return response.status == 200 + except (OSError, urllib.error.URLError): + return False + with tempfile.TemporaryDirectory(prefix='dsh-lifecycle-') as temporary, smoke.MockModel() as model: + cwd = Path(temporary) + with DeepSeekHarness(provider='deepseek-official', model='smoke-model', + profile='sdk-minimal', cwd=str(cwd), dsh_home=str(cwd / 'home'), + dsh_bin=str(runtime), api_key='keyless-lifecycle-probe', + base_url=model.url, request_timeout_seconds=60) as harness: + result = harness.run('Start the probe server.', session_id='lifecycle') + before_close = reachable() + time.sleep(0.25) + after_close = reachable() + # The probe must not leave its explicitly created HTTP server behind. + if after_close: + import os + import signal + pid = int((cwd / 'server.pid').read_text()) + command = (Path('/proc') / str(pid) / 'cmdline').read_bytes() + if b'http.server' in command and str(port).encode() in command: + os.kill(pid, signal.SIGTERM) + report = { + 'checked_at': datetime.now(timezone.utc).isoformat(), + 'profile': 'sdk-minimal', 'finish_reason': result.finish_reason, + 'local_mock_requests': calls, 'real_model_calls': 0, + 'server_reachable_before_sdk_close': before_close, + 'server_reachable_after_sdk_close': after_close, + 'probe_sha256': hashlib.sha256(Path(__file__).read_bytes()).hexdigest(), + 'upstream_fixture_sha256': hashlib.sha256(fixture.read_bytes()).hexdigest(), + 'runtime_sha256': hashlib.sha256(runtime.read_bytes()).hexdigest(), + } + print(json.dumps(report, indent=2)) + return 0 if result.finish_reason == 'completed' and before_close and calls == 2 else 1 + + +if __name__ == '__main__': + raise SystemExit(main()) From eff073b1601435305155cad92b6ee9e2250fbbdc Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 06:35:23 -0800 Subject: [PATCH 08/15] feat(benchmarks): preflight dependencies for the deepseek supplement --- benchmarks/harbor/README.md | 9 +++ benchmarks/harbor/README.zh-CN.md | 7 +++ .../harbor/src/harbor_adapter/bootstrap.py | 58 +++++++++++++++++++ benchmarks/harbor/tests/test_bootstrap.py | 38 ++++++++++++ 4 files changed, 112 insertions(+) diff --git a/benchmarks/harbor/README.md b/benchmarks/harbor/README.md index 915d1c1..d34cd27 100644 --- a/benchmarks/harbor/README.md +++ b/benchmarks/harbor/README.md @@ -88,6 +88,15 @@ pytest 8.4.1, torch 2.7.1, and pytest-json-ctrf 0.3.5 requirements, then runs on `pytest --version` from `/tmp`. It does not import the submitted model or run the task's tests. Other revisions require profile review. +Reviewed revisions of `multi-source-data-merger`, `pytorch-model-cli`, +`query-optimize`, `torch-pipeline-parallelism`, and `torch-tensor-parallelism` +also have dependency-only profiles. They preserve each original wrapper's +Python version, package pins, APT dependencies, and index options, and run only +`pytest --version` from `/tmp`. In particular, only the CLI profile adds the +original PyTorch CPU index and `unsafe-best-match` strategy; the other PyTorch +profiles keep their original default index. Task metadata and verifier-wrapper +hashes are recorded with these five profiles. + Every installation writes `agent/bootstrap.json`, including partial progress on failure. Preflight failure stops the trial before any model call. The official verifier script remains unchanged and may still access the network diff --git a/benchmarks/harbor/README.zh-CN.md b/benchmarks/harbor/README.zh-CN.md index a2d1386..e21030f 100644 --- a/benchmarks/harbor/README.zh-CN.md +++ b/benchmarks/harbor/README.zh-CN.md @@ -71,6 +71,13 @@ uv 0.9.5、Python 3.13、pytest 8.4.1 和 pytest-json-ctrf 0.3.5,不执行题 0.3.5,随后仅在 `/tmp` 执行 `pytest --version`。它不导入提交的模型,也不运行 题目测试。其他题目版本需要重新检查对应配置。 +已检查的 `multi-source-data-merger`、`pytorch-model-cli`、`query-optimize`、 +`torch-pipeline-parallelism` 和 `torch-tensor-parallelism` 版本也有仅准备依赖的 +配置。它们保留各自原始脚本的 Python 版本、软件包版本、APT 依赖和索引选项, +仅在 `/tmp` 运行 `pytest --version`。其中仅 CLI 配置添加原脚本中的 PyTorch +CPU 索引和 `unsafe-best-match` 策略;其他 PyTorch 配置保留原始默认索引。 +每个新检查的配置都记录题目元数据与验证器脚本的哈希。 + 每次安装都会写入 `agent/bootstrap.json`,失败时也保留已完成的步骤。预检失败会 在模型调用前终止 trial。官方验证脚本保持原样,执行时仍可能访问网络,所以预检 通过不能保证后续没有网络故障。完整标准运行默认启用下述验证超时重试,整题重试 diff --git a/benchmarks/harbor/src/harbor_adapter/bootstrap.py b/benchmarks/harbor/src/harbor_adapter/bootstrap.py index 5a0f550..858b62d 100644 --- a/benchmarks/harbor/src/harbor_adapter/bootstrap.py +++ b/benchmarks/harbor/src/harbor_adapter/bootstrap.py @@ -165,6 +165,21 @@ class VerifierProfile: task_files_sha256: tuple[tuple[str, str], ...] = () +def python_verifier_command(requirements: tuple[str, ...], *, indexes=()) -> str: + """Warm a reviewed uvx environment without importing task code or tests.""" + return ( + "set -euo pipefail; cd /tmp; command -v curl; " + "curl -LsSf https://astral.sh/uv/0.9.5/install.sh | sh; " + 'export PATH="$HOME/.local/bin:$PATH"; ' + "uv --version | grep -Eq '^uv 0\\.9\\.5( |$)'; " + + shlex.join([ + "uvx", *indexes, "-p", "3.13", + *[arg for requirement in requirements for arg in ("-w", requirement)], + "pytest", "--version", + ]) + ) + + VERIFIER_PROFILES = { (QEMU_TASK, QEMU_REF): VerifierProfile( "qemu-alpine-ssh-v1", ("curl", "sshpass"), QEMU_PREFLIGHT, @@ -174,6 +189,49 @@ class VerifierProfile: (("task.toml", "851a51204f53482a31f47cdf8833128cf421113b0331011286940fc857c37dfd"), ("tests/test.sh", "6005f3a75ccfe00d7bed1005264b4a84a1f8e30cd780f8a97b3a804be7cc7a8e")), ), + ("terminal-bench/multi-source-data-merger", + "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49"): VerifierProfile( + "multi-source-data-merger-v1", ("curl",), python_verifier_command(( + "pytest==8.4.1", "pandas==2.3.3", "pyarrow==22.0.0", "pytest-json-ctrf==0.3.5", + )), + (("task.toml", "ee3964c0ee040f0a0cb121cefefbc15443d127684acb003fa5b61dd42f31c694"), + ("tests/test.sh", "d5a00e64413c0e1ae452f38a1aa1a47c3ed7b4416bbfccae1a4c053025a336d1")), + ), + ("terminal-bench/pytorch-model-cli", + "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79"): VerifierProfile( + "pytorch-model-cli-v1", ("curl", "ffmpeg", "libsm6", "libxext6"), + python_verifier_command(( + "torch==2.7.1", "torchvision==0.22.1", "numpy==2.3.1", + "opencv-python==4.11.0.86", "pytest==8.4.1", "pytest-json-ctrf==0.3.5", + ), indexes=("--index", "https://download.pytorch.org/whl/cpu", + "--index-strategy", "unsafe-best-match")), + (("task.toml", "fc1c018efbd5d0537ac0741d53d1ebad3886faa51d14cbde8ec953856411bc0d"), + ("tests/test.sh", "43fb67e537c7adc2df11c2367f47c876cb5d16a7c1ae16d0339c3a3364bb2b06")), + ), + ("terminal-bench/query-optimize", + "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4"): VerifierProfile( + "query-optimize-v1", ("curl",), python_verifier_command(( + "pytest==8.4.1", "pytest-json-ctrf==0.3.5", + )), + (("task.toml", "6e88f2a86df699594bcb1bf81df4ed694505f79c9420f671335b02524bbe3868"), + ("tests/test.sh", "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334")), + ), + ("terminal-bench/torch-pipeline-parallelism", + "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb"): VerifierProfile( + "torch-pipeline-parallelism-v1", ("curl",), python_verifier_command(( + "pytest==8.4.1", "torch==2.7.0", "transformers==4.55.0", "pytest-json-ctrf==0.3.5", + )), + (("task.toml", "8b4f9f525c08afbbc4eca9168e46238a61ccb903486773e9ca4da64bb7ffe857"), + ("tests/test.sh", "cc9fd28b6590993fc37e4a99da771cbf04726c8c62fe308e802612241da69dea")), + ), + ("terminal-bench/torch-tensor-parallelism", + "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684"): VerifierProfile( + "torch-tensor-parallelism-v1", ("curl",), python_verifier_command(( + "pytest==8.4.1", "torch==2.7.0", "pytest-json-ctrf==0.3.5", + )), + (("task.toml", "72fd2a5500d1fe8795dcae76cccb1a29fb25894881a7475f2fa519c87d425a8d"), + ("tests/test.sh", "a7e6c9408b2401e4e7448371f7ddc72f3284caccfb56277d3cd4e98051868cdf")), + ), } diff --git a/benchmarks/harbor/tests/test_bootstrap.py b/benchmarks/harbor/tests/test_bootstrap.py index 882ead5..fcac335 100644 --- a/benchmarks/harbor/tests/test_bootstrap.py +++ b/benchmarks/harbor/tests/test_bootstrap.py @@ -5,6 +5,7 @@ import hashlib import json from pathlib import Path +import shlex from types import SimpleNamespace from unittest.mock import AsyncMock @@ -151,6 +152,43 @@ def test_pytorch_recovery_profile_pins_the_original_verifier_dependencies(tmp_pa assert set(dict(profile.task_files_sha256)) == {"task.toml", "tests/test.sh"} +@pytest.mark.parametrize("name,requirements,apt_packages", [ + ("multi-source-data-merger", ["pytest==8.4.1", "pandas==2.3.3", "pyarrow==22.0.0", "pytest-json-ctrf==0.3.5"], ("curl",)), + ("pytorch-model-cli", ["torch==2.7.1", "torchvision==0.22.1", "numpy==2.3.1", "opencv-python==4.11.0.86", "pytest==8.4.1", "pytest-json-ctrf==0.3.5"], ("curl", "ffmpeg", "libsm6", "libxext6")), + ("query-optimize", ["pytest==8.4.1", "pytest-json-ctrf==0.3.5"], ("curl",)), + ("torch-pipeline-parallelism", ["pytest==8.4.1", "torch==2.7.0", "transformers==4.55.0", "pytest-json-ctrf==0.3.5"], ("curl",)), + ("torch-tensor-parallelism", ["pytest==8.4.1", "torch==2.7.0", "pytest-json-ctrf==0.3.5"], ("curl",)), +]) +def test_supplement_profiles_preserve_reviewed_dependencies_and_fail_closed( + tmp_path, name, requirements, apt_packages, +): + # The original registration independently pins the reviewed task files. + registration = Path(__file__).resolve().parents[1] / "configs/deepseek-harness-dev34-20261030/experiment.json" + task = next(t for t in json.loads(registration.read_text())["tasks"] + if t["name"] == "terminal-bench/" + name) + config = tmp_path / "config.json" + config.write_text(json.dumps({"task": {"name": task["name"], "ref": task["ref"]}})) + profile = verifier_profile(tmp_path / "agent", name) + assert profile.apt_packages == apt_packages + assert dict(profile.task_files_sha256) == { + file: task["task_file_sha256"][file] for file in ("task.toml", "tests/test.sh") + } + args = shlex.split(profile.command.rsplit("; ", 1)[1]) + assert [args[i + 1] for i, arg in enumerate(args) if arg == "-w"] == requirements + assert args[-2:] == ["pytest", "--version"] + assert args[args.index("-p") + 1] == "3.13" + assert "cd /tmp" in profile.command and "uv/0.9.5/install.sh" in profile.command + assert "/tests" not in profile.command and "/app" not in profile.command + if name == "pytorch-model-cli": + assert args[1:5] == ["--index", "https://download.pytorch.org/whl/cpu", + "--index-strategy", "unsafe-best-match"] + else: + assert not any(arg.startswith("--index") for arg in args) + config.write_text(json.dumps({"task": {"name": task["name"], "ref": "latest"}})) + with pytest.raises(ValueError, match="Unknown .*revision"): + verifier_profile(tmp_path / "agent", name) + + @pytest.mark.parametrize("task", [ {"path": "/tasks/qemu-alpine-ssh"}, {"path": "/dataset/qemu-alpine-ssh", "source": "local-dataset"}, From 508f66c7bacfec67053b2ae41863d8a57507cbc3 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 06:37:15 -0800 Subject: [PATCH 09/15] test(benchmarks): register the authorized six-task deepseek supplement --- .../README.md | 69 ++++ .../apt-cache.conf | 8 + .../deepseek-distribution.json | 58 ++++ .../dependency-cache.compose.json | 101 ++++++ .../dependency-release-manifest.json | 58 ++++ .../experiment.json | 320 ++++++++++++++++++ .../image-digests.json | 60 ++++ .../install-pair-01.json | 42 +++ .../install-pair-02.json | 42 +++ .../install-pair-03.json | 42 +++ .../model-alias-check.json | 9 + .../multi-source-data-merger.json | 38 +++ .../pytorch-model-cli.json | 38 +++ .../pytorch-model-recovery.json | 38 +++ .../query-optimize.json | 38 +++ .../selection.json | 13 + .../torch-pipeline-parallelism.json | 38 +++ .../torch-tensor-parallelism.json | 38 +++ 18 files changed, 1050 insertions(+) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md new file mode 100644 index 0000000..3d1b069 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md @@ -0,0 +1,69 @@ +# Authorized six-task DeepSeek Harness supplement + +Status: registered on November 1, 2026; dependency preflight pending. No +supplemental model executions have started. + +The maintainer explicitly requested another run of the six tasks left +analytically unscored in the [original dev34 results](../../results/deepseek-harness-dev34-20261030.json). +This is a separate, outcome-selected development supplement. Keep the original +22 passes, six failures, and six unscored errors unchanged. Record every new +attempt separately, including preparation failures; never choose the better +of two scores or present the supplement as an independent clean benchmark. + +The six selected tasks are `multi-source-data-merger`, `pytorch-model-cli`, +`pytorch-model-recovery`, `query-optimize`, `torch-pipeline-parallelism`, and +`torch-tensor-parallelism`, in their original relative order. Task revisions, +files, images, official test scripts, resources, and native time limits are +unchanged. The other 55 tasks remain unopened. Earlier failures and diagnostic +material are not included in model prompts. + +The upstream Harness distribution is unchanged: source +`f3803c96974ad4a94195fb97e5c637e9c2a3afe3`, `sdk-minimal`, native prompt/tools, +compaction and request retries, `deepseek/deepseek-flash`, high reasoning, +Messages endpoint, temperature and SDK max_tokens unset. The official guide +checked on November 1 still documents DeepSeek-V4.1-Flash-1017 for this mutable +alias; this does not establish an identical backend checkpoint across dates. +No service-lifecycle change is included. + +The frozen adapter source is `eff073b1601435305155cad92b6ee9e2250fbbdc`, at +`jobs/deepseek-harness-supplement-source`. Relative to the original experiment, +it adds reviewed dependency-only preflight profiles for the five previously +unsupported tasks in this subset. Model recovery retains its existing profile. +The commands preserve the original Python version, package pins, APT +requirements, and index options, and run only `pytest --version` from `/tmp`. +All 285 Harbor tests passed, including regressions for those profiles. + +The task environment sets `UV_HTTP_TIMEOUT=120` seconds instead of the default +30 to tolerate download stalls. It retains original package sources and the +existing verified APT-index and uv/Python-release caches. This transport change +does not increase the task-native agent/verifier limits. Setup remains 1,800 +seconds; concurrency remains two with pair barriers. There is one fresh model +attempt per selected task and no automatic whole-trial retry. Only verifier +timeouts may retry, at most three total attempts on the same new output. + +All six fresh install-only checks, including their reviewed dependency +profiles, must pass before model work. Every formal job repeats dependency +setup before calling the model. Failure at that point is a setup error, not a +scored failure. Preserve all new outputs, freeze them before reviewing new +trajectories or hidden assertions, and publish full English and Simplified +Chinese reports with original and supplemental outcomes distinguished. + +```bash +export UV_PROJECT_ENVIRONMENT="$PWD/jobs/dsh-harbor-venv" +export UV_CACHE_DIR="$PWD/.cache/uv" +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json \ + --phase preflight +# Supply the previously authorized DEEPSEEK_API_KEY in the environment. +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json \ + --phase model +``` + +Every job uses the standard `python -m harbor_adapter run --config ...` entry +point from the shared frozen source. Native outputs remain under +`jobs/deepseek-harness-supplement-source/jobs/`; orchestration and image events +remain under `jobs/deepseek-harness-dev6-supplement-20261101-orchestration/`. +The existing original registration and frozen outputs are not overwritten. diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf new file mode 100644 index 0000000..e191ddb --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf @@ -0,0 +1,8 @@ +// Route only public APT HTTP origins through the verified index cache. +Acquire::http::Proxy::deb.debian.org "http://127.0.0.1:18742/"; +Acquire::http::Proxy::security.debian.org "http://127.0.0.1:18742/"; +Acquire::http::Proxy::archive.debian.org "http://127.0.0.1:18742/"; +Acquire::http::Proxy::archive.ubuntu.com "http://127.0.0.1:18742/"; +Acquire::http::Proxy::security.ubuntu.com "http://127.0.0.1:18742/"; +Acquire::http::Proxy::ports.ubuntu.com "http://127.0.0.1:18742/"; +Acquire::http::Proxy::old-releases.ubuntu.com "http://127.0.0.1:18742/"; diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json new file mode 100644 index 0000000..8b38172 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json @@ -0,0 +1,58 @@ +{ + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "provenance": { + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "source_archive_sha256": "0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4", + "origin": "https://github.com/deepseek-ai/deepseek-harness.git", + "node_version": "26.7.0", + "pnpm_version": "11.28.5", + "dependency_install": "pnpm install --frozen-lockfile --store-dir /mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/dsh-pnpm", + "runtime_build": "DSH_GATE_CONCURRENCY=4 pnpm exec tsx scripts/build-exe-for-python-sdk.ts --targets=node24-linux-x64 --build-artifacts --compress=GZip", + "source_git_metadata": "Original commit objects imported from the local source checkout after the official build required git rev-parse HEAD; source files remain unchanged.", + "source_clean_after_build": true, + "build_log_sha256": "73ad636ed60974c115c5d4dd0bc3819779908fb0f6cf75c95fe7e834e5838198", + "adapter_ref": "65d0fd389c61fc3a98d150be789c33da39e36f6a", + "sdk_minimal_smoke": { + "status": "passed", + "log_sha256": "df510602b41bd2c0a7aa83ed4867d8a5b675b43b2e5cc1748f7b1640b94bb9b4" + } + }, + "files": { + "build-provenance.json": { + "sha256": "36212ebe23e42e7cfab847266bac7a4f8a05d83bbce4e5422f7dcfd2f87d4785", + "mode": 420 + }, + "distribution.json": { + "sha256": "c8d7a7842b835f66eb6dd0230403f0b4af0a26dd0e05ea04810c9667d7ba3d57", + "mode": 420 + }, + "wheels/annotated_types-0.8.0-py3-none-any.whl": { + "sha256": "f072f4d804ea359e4eaf198b1af7a8b0943881a87f31bb764f8bf219bb9419e0", + "mode": 420 + }, + "wheels/deepseek_harness_runtime_bin-0.2.2a2-py3-none-manylinux_2_28_x86_64.whl": { + "sha256": "1fb53195ac1c6d5bee02a63f11488487df74d964ee7fb1603eb268d8116edddc", + "mode": 420 + }, + "wheels/deepseek_harness_sdk-0.2.2a2-py3-none-any.whl": { + "sha256": "282bfb4b373e8b4634447bddc066f6cc0f20da5d422ed5618f37d469823b742f", + "mode": 420 + }, + "wheels/pydantic-2.13.4-py3-none-any.whl": { + "sha256": "45a282cde31d808236fd7ea9d919b128653c8b38b393d1c4ab335c62924d9aba", + "mode": 420 + }, + "wheels/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl": { + "sha256": "9551187363ffc0de2a00b2e47c25aeaeb1020b69b668762966df15fc5659dd5a", + "mode": 420 + }, + "wheels/typing_extensions-4.16.0-py3-none-any.whl": { + "sha256": "481caa481374e813c1b176ada14e97f1f67a4539ce9cfeb3f350d78d6370c2e8", + "mode": 420 + }, + "wheels/typing_inspection-0.4.4-py3-none-any.whl": { + "sha256": "65b8397ba37ccbce054456aaccddfc91e6e3083c92824df348d96ca832f3f147", + "mode": 420 + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json new file mode 100644 index 0000000..4c51733 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json @@ -0,0 +1,101 @@ +{ + "services": { + "main": { + "volumes": [ + { + "type": "bind", + "source": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf", + "target": "/etc/apt/apt.conf.d/99nanopy-index-cache", + "read_only": true + } + ], + "environment": { + "UV_INSTALLER_GITHUB_BASE_URL": "http://127.0.0.1:18743", + "UV_PYTHON_INSTALL_MIRROR": "http://127.0.0.1:18743/astral-sh/python-build-standalone/releases/download", + "UV_HTTP_TIMEOUT": "120" + } + }, + "apt-index-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/apt_index_proxy.py", + "--bind", + "127.0.0.1", + "--port", + "18742", + "--cache", + "/cache", + "--workers", + "16" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "mem_limit": "256m", + "cpus": 0.5, + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/apt_index_proxy.py:/proxy/apt_index_proxy.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/apt-indexes:/cache" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18742), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + }, + "uv-release-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/uv_release_cache.py", + "serve", + "--cache", + "/cache", + "--manifest", + "/manifest.json" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "cpus": 0.5, + "mem_limit": "256m", + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/uv_release_cache.py:/proxy/uv_release_cache.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/uv-releases:/cache:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json:/manifest.json:ro" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18743), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json new file mode 100644 index 0000000..e762efb --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json @@ -0,0 +1,58 @@ +{ + "registered_at": "2026-10-18T13:25:43.131474+00:00", + "metadata_source": "Original release metadata preserved by the source manifests; no versions or payloads substituted.", + "source_manifests": { + "benchmarks/harbor/configs/dev34-main-baseline-20261012/dependency-release-manifest-v5.json": "bdf3d3ef8b7743700c5294c4a511c686a3928c178a77fd4abcba90ee4557ffcc", + "python-agent-release-manifest.json": "78aecd7c397c469196f9d246a160bf1fc99ad2d51cb71320b7d26ebf839647ce" + }, + "assets": [ + { + "version": "0.9.11", + "filename": "uv-x86_64-unknown-linux-gnu.tar.gz", + "url": "https://github.com/astral-sh/uv/releases/download/0.9.11/uv-x86_64-unknown-linux-gnu.tar.gz", + "digest": "sha256:817c0722b437b4b45b9a7e0231616a09db76bab1b8d178ba7a9680c690db19f0", + "size": 21699161, + "asset_id": 318993196, + "release_metadata_sha256": "c1a1378ece65ce1aa8343e57bdbd305289a8fc6d3d9180e8bb6b0f69acd91413" + }, + { + "version": "0.9.5", + "filename": "uv-x86_64-unknown-linux-gnu.tar.gz", + "url": "https://github.com/astral-sh/uv/releases/download/0.9.5/uv-x86_64-unknown-linux-gnu.tar.gz", + "digest": "sha256:2cf10babba653310606f8b49876cfb679928669e7ddaa1fb41fb00ce73e64f66", + "size": 21370871, + "asset_id": 306970260, + "release_metadata_sha256": "1921b17d3370aafb690b29d1d121ee8719ef3d3dbc0c7cf3b948c3059a927a5d" + }, + { + "version": "20251014", + "filename": "cpython-3.13.9+20251014-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "url": "https://github.com/astral-sh/python-build-standalone/releases/download/20251014/cpython-3.13.9%2B20251014-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "digest": "sha256:c0ccda275948c79996e46993c2c5476ff5cca606dee530f1dea712179131b348", + "size": 33517986, + "asset_id": 304305340, + "release_metadata_sha256": "8e5db21cfcc2bcafcb2127f269f00cbfdd1f74842a685acd2a14e3c1ad6f8394", + "repository": "astral-sh/python-build-standalone" + }, + { + "version": "20251120", + "filename": "cpython-3.14.0+20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "url": "https://github.com/astral-sh/python-build-standalone/releases/download/20251120/cpython-3.14.0%2B20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "digest": "sha256:66f385305ae0eefd6b65e2cf942bc91d943a61e26fb7581e029f760a2b04f393", + "size": 35376131, + "asset_id": 318881850, + "release_metadata_sha256": "e3a737862d2dddf11c73bfef6c2f2579e9d4ce5bd271cca53790c998590dee58", + "repository": "astral-sh/python-build-standalone" + }, + { + "version": "20251120", + "filename": "cpython-3.13.9+20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "url": "https://github.com/astral-sh/python-build-standalone/releases/download/20251120/cpython-3.13.9%2B20251120-x86_64-unknown-linux-gnu-install_only_stripped.tar.gz", + "digest": "sha256:1bc9a94ab2425fb6ae58e45f69cab61647759b57895649701390822d976fa923", + "size": 34257108, + "asset_id": 318881608, + "release_metadata_sha256": "705690ab2fd466e90b4c868635c529f3c7fd0cb597d399e350f399e3f3e35e6c", + "repository": "astral-sh/python-build-standalone" + } + ] +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json new file mode 100644 index 0000000..a969f29 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json @@ -0,0 +1,320 @@ +{ + "schema_version": 1, + "registered_at": "2026-11-01T14:36:34.471637+00:00", + "experiment": "deepseek-harness-dev6-supplement-20261101", + "total_model_runs": 6, + "max_concurrent_trials": 2, + "whole_trial_retries": 0, + "minimum_free_disk_bytes": 16106127360, + "log_dir": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration", + "arms": { + "deepseek": { + "agent": "deepseek-harness", + "source_root": "jobs/deepseek-harness-supplement-source", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6" + } + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/multi-source-data-merger:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-processing", + "verifier_dependency_profile": "multi-source-data-merger-v1", + "task_file_sha256": { + "README.md": "9bf37162a4c1b88ed93627830bb8512ebae154325d7cb069c883144a0e4c69d6", + "environment/Dockerfile": "c95a40d3d35b473dc15ae1e8784ba7fb1725805c0583fc574dd2764325d75e4f", + "environment/data/source_a/users.json": "75e51e2d3ab0bdba2269d7cf3757576f84f3ee2c13407d2db94289a5958a1778", + "environment/data/source_b/users.csv": "8e87d16dc47d166edd8e56165955bad265b630d53d02762367c816affb5fa760", + "environment/data/source_c/users.parquet": "23d1ea51c4170bdba53febb885a0f495e28de525a52667347b2932136b9a72fe", + "instruction.md": "d1d28bb6e37f7a255a1b7f78aae8591f35c0494f33e4913392424ad954aab944", + "solution/solve.sh": "1b3d0d50e31930d042495d48f4c8e95f1a9c1a979c803b7f7cec651322537e1e", + "task.toml": "ee3964c0ee040f0a0cb121cefefbc15443d127684acb003fa5b61dd42f31c694", + "tests/test.sh": "d5a00e64413c0e1ae452f38a1aa1a47c3ed7b4416bbfccae1a4c053025a336d1", + "tests/test_outputs.py": "a070888930cf829aeca039771260b1386bda990ea805f313468d381ba3608ca5" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pytorch-model-cli:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "model-training", + "verifier_dependency_profile": "pytorch-model-cli-v1", + "task_file_sha256": { + "README.md": "35a093fa5f1592737784ba718c2e3e9ae17ff8b5e53e003fae560ddce091039c", + "environment/Dockerfile": "8fd5c0f21a6914cd440d10c84c9db06fec606f20dc3e0363b17a3b8c220dc4d9", + "environment/task-deps/cJSON.c": "0e7995261640c96db16ed2c91a5d981efa917cf5e72d0f97ec45a1179ae3ad73", + "environment/task-deps/cJSON.h": "0909160541bcc510e0b2d8c7a9cc051fa1459a0f13be4d8e56346a15491f2f2f", + "environment/task-deps/image.png": "f4d5dfefe463939996541404340e7099d5957b56e1df3760d13b5d2de754db53", + "environment/task-deps/lodepng.cpp": "67873c5f6272f779a143a9c81070e8783327c5552d0b9de2aaee2cb779308917", + "environment/task-deps/lodepng.h": "b735b7e0357319973e9488d520d7f270d734a6e9e1bf924fb1d5b8d26d7e97ce", + "environment/task-deps/model.pth": "a32103eef09936e586b1a7dc7e64d279500a38ff30fdd6df2167e28eb954eee6", + "environment/task-deps/model.py": "e4d3a5c728449656d0efde7871ee227619348352769795ed1be1a8b7f835fddc", + "instruction.md": "ce36ee1f6c589f81ee6a6be4967e25c76a2ed697006928023f42dfb9f28a6750", + "solution/solve.sh": "3b34e561c9ef0a53c9948655847846eb4811aaac67b990480aa6aac1b04bfe66", + "task.toml": "fc1c018efbd5d0537ac0741d53d1ebad3886faa51d14cbde8ec953856411bc0d", + "tests/test.sh": "43fb67e537c7adc2df11c2367f47c876cb5d16a7c1ae16d0339c3a3364bb2b06", + "tests/test_outputs.py": "370702c57a4e0c85f70ee37a5c357ae3a5df8480a275bb6263b0376f2f6c093b" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pytorch-model-recovery:20260430", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "model-training", + "verifier_dependency_profile": "pytorch-model-recovery-v1", + "task_file_sha256": { + "README.md": "e080308a05703b40edd16f53aea41a0e6b9874750ecee35bb432029d0d44bf53", + "environment/Dockerfile": "8a6b7edbb9e7c241e447f15a71e13e5d043b25169be2fd9802b84f5be3d56eb6", + "environment/dataset.pt": "6db65e4ec1c0760209c2dd997ef582e4a1519ab38e0312339999c460a696960d", + "environment/weights.pt": "71471e71c5a311a897362dd5b77413e7ca2ca59abd00a110bdc2cd3f8584d080", + "instruction.md": "2bef07dc28be3ab2c35ae389584034c2e4dde92c2c31b3eb6bd829628c8c3f4a", + "solution/solve.sh": "175a7802777b43e0fa4f629d71f7c68fb40706813c342aaddc72108d89bd5b2c", + "task.toml": "851a51204f53482a31f47cdf8833128cf421113b0331011286940fc857c37dfd", + "tests/test.sh": "6005f3a75ccfe00d7bed1005264b4a84a1f8e30cd780f8a97b3a804be7cc7a8e", + "tests/test_outputs.py": "5d4d57c43e6913bf3d0081da764ad36d1b7275e3bb114a05abda529afe563905", + "tests/weights_gtruth.pt": "71471e71c5a311a897362dd5b77413e7ca2ca59abd00a110bdc2cd3f8584d080" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/query-optimize:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-science", + "verifier_dependency_profile": "query-optimize-v1", + "task_file_sha256": { + "README.md": "a82c53e333a0ac0b0b410b5e26cb8e7305dece31db84d340248ffb13d5a45160", + "environment/Dockerfile": "064818771f200e0517d22068f87229a7a95d0a82b0778c84ca7203bf1926a0b8", + "environment/task-deps/my-sql-query.sql": "43382792f0f438625d643e34abe144febd5669518f4ab1ab550a15c6d9314a21", + "instruction.md": "473e12d320d2732f679fb631c2a4dab4a01deea6775686718ae0c417b8a39447", + "solution/solve.sh": "0d95f3cedb287ff9d47c796cbaf812c300d1222c10fa80c49e8acb0e81eb6978", + "task.toml": "6e88f2a86df699594bcb1bf81df4ed694505f79c9420f671335b02524bbe3868", + "tests/golden.sql": "4da1885d56b3ffa7b7b22b422996d018795611f6963499d551582881b74b84fd", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "2068ae80c57af8d52a7ae510f0666732c057806e556b480970f36f14bae683f6" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/torch-pipeline-parallelism:20251031", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "torch-pipeline-parallelism-v1", + "task_file_sha256": { + "README.md": "297ddea07c230dfd246a8ab07337dd7abcd8e5e2a2b2e26590cb0a8af779839c", + "environment/Dockerfile": "c357ea41e4a4013c5c5980b8fef04af6a5ae250ccc6697cf3bad3127401a931f", + "instruction.md": "989bfcf0c54baf6c8cdc391743dbb2f9e858c5779dd85112536875d4d1b2dcd2", + "solution/solve.sh": "787198e4178d88b887697aefdd29819f29435ffdfe356545d92f58a30fb38b70", + "task.toml": "8b4f9f525c08afbbc4eca9168e46238a61ccb903486773e9ca4da64bb7ffe857", + "tests/test.sh": "cc9fd28b6590993fc37e4a99da771cbf04726c8c62fe308e802612241da69dea", + "tests/test_outputs.py": "23574246a407163a8e6581c84287cc2f1eb745efbaad2a317a5a28c39958c64f" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/torch-tensor-parallelism:20251031", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "torch-tensor-parallelism-v1", + "task_file_sha256": { + "README.md": "4c1b3c3be30548f1f94c6f5503b1bc74b006af7f192dc4cd026f411df1a54604", + "environment/Dockerfile": "5f264ff0bf62a49aee3c6398447de5108df141e7a2ad35c6de8ef39fa62049c6", + "instruction.md": "13bde2ab165d72a311e2bf066d5d6c3c8761bf9d3dd39d1b23eda23bc546b910", + "solution/solve.sh": "d083b7cbf9dbb6cc2ff38c8f2acc150f0059c04aea42d6c2183aeaecc5a19385", + "task.toml": "72fd2a5500d1fe8795dcae76cccb1a29fb25894881a7475f2fa519c87d425a8d", + "tests/test.sh": "a7e6c9408b2401e4e7448371f7ddc72f3284caccfb56277d3cd4e98051868cdf", + "tests/test_outputs.py": "4ae4285e15fba1fcf13d7b7c0638ac8c2e2b0797c937abab1fb6b136fd508290" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + } + ], + "preflight": [ + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json", + "config_sha256": "cc4f07cff8a7ec9ffe4de4bdb69322b0ec89b62c72072020d1f9b1f7f0e20e26", + "job_name": "dsh-sup6-install-pair-01-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json", + "config_sha256": "8c918a0998feb41ac3a974228d1bc3b4ba220fb5d76283c77590755738c1d149", + "job_name": "dsh-sup6-install-pair-02-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json", + "config_sha256": "b04b132fa453bec7c4f888e6e52dab533b8870dc1c0ea21881fb887c84ac00ba", + "job_name": "dsh-sup6-install-pair-03-20261101" + } + ], + "queue": [ + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json", + "config_sha256": "4e25ecef8ce9beb5a0adfb9977242fe32a9a5559b8f691fa686971d4a6d4e4e1", + "task": "terminal-bench/multi-source-data-merger", + "repeat": 1, + "job_name": "dsh-sup6-multi-source-data-merger-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json", + "config_sha256": "a0805861bc826d29e2bc9205c2c720f725f52a549d970c61b407f786dd73df49", + "task": "terminal-bench/pytorch-model-cli", + "repeat": 1, + "job_name": "dsh-sup6-pytorch-model-cli-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json", + "config_sha256": "fddffa4e9c5bac957f87460cdd04276f964fb4c2aa1bcd98124387d95f140c99", + "task": "terminal-bench/pytorch-model-recovery", + "repeat": 1, + "job_name": "dsh-sup6-pytorch-model-recovery-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json", + "config_sha256": "2320f05a860f4c91d1fd1a200740cda5ac8fa67ea65dfab3dc3a5b0ef745b6e0", + "task": "terminal-bench/query-optimize", + "repeat": 1, + "job_name": "dsh-sup6-query-optimize-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json", + "config_sha256": "bfa979780e67235b1f5472d025ca7b8cb2ea2b4ef2c7f1fab29d5b4fabee8d5c", + "task": "terminal-bench/torch-pipeline-parallelism", + "repeat": 1, + "job_name": "dsh-sup6-torch-pipeline-parallelism-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json", + "config_sha256": "11b91bacfe4d1b2fdcf57a23376ae61b16df9c69802bd5465557db95b95976d8", + "task": "terminal-bench/torch-tensor-parallelism", + "repeat": 1, + "job_name": "dsh-sup6-torch-tensor-parallelism-20261101" + } + ], + "additional_file_sha256": { + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf": "d0e1ffb6dbd02c54978ba2df386326311a59f6d3774990180f22bcc6adfaedbc", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json": "34410bdce15ec3b417105eb7f9f6e01a14c47f1bfd0ffe16d76338e7efa51dd2", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json": "560bc83f2a88bb907662db1b60ab15f49443a1572822fa633440f3cc76510557", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json": "77dfda40e45b870a8ce7f07a1e058d416d94a04ab3d27c2f21bc370f22151394", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json": "c2405e16c2aaf2d837b6f6594875e197fd13eb94ed094253e4513c315783d030", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json": "cc4f07cff8a7ec9ffe4de4bdb69322b0ec89b62c72072020d1f9b1f7f0e20e26", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json": "8c918a0998feb41ac3a974228d1bc3b4ba220fb5d76283c77590755738c1d149", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json": "b04b132fa453bec7c4f888e6e52dab533b8870dc1c0ea21881fb887c84ac00ba", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json": "452c5114c9ac5fc82634ce25ce8a2ee0a8f4e3ee15724f5324bdb1befcf865c3", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json": "4e25ecef8ce9beb5a0adfb9977242fe32a9a5559b8f691fa686971d4a6d4e4e1", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json": "a0805861bc826d29e2bc9205c2c720f725f52a549d970c61b407f786dd73df49", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json": "fddffa4e9c5bac957f87460cdd04276f964fb4c2aa1bcd98124387d95f140c99", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json": "2320f05a860f4c91d1fd1a200740cda5ac8fa67ea65dfab3dc3a5b0ef745b6e0", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json": "5047e6d007ed45ee6887bcf09cceb3475b7075e56551a43143c38b8e00991284", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json": "bfa979780e67235b1f5472d025ca7b8cb2ea2b4ef2c7f1fab29d5b4fabee8d5c", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json": "11b91bacfe4d1b2fdcf57a23376ae61b16df9c69802bd5465557db95b95976d8", + "benchmarks/harbor/scripts/run_comparison.py": "228dbff33c0ec8a2359e612da632ef543dcac4273db2f9a7d0573a24e856fbae", + "benchmarks/harbor/scripts/apt_index_proxy.py": "bc15f7adf1d7968d627ceca4683ab86742a7188a1b1955ba1fef37ba48e8a936", + "benchmarks/harbor/scripts/cache_image.py": "11ddde3fe86a6e4bfdefe7c3578794df98d237f701e40f219d76a541911c9f0c", + "benchmarks/harbor/scripts/uv_release_cache.py": "993cc35b1f3a9e8e171b662ab6d4524c5fe059f0d1f0e275a5bb1930111e0ff5", + "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396" + }, + "batch_images": { + "manifest": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json", + "storage_path": "/var/lib/docker", + "minimum_free_disk_bytes": 16106127360, + "cache_dir": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/images" + }, + "model_policy": { + "model": "deepseek/deepseek-flash", + "provider": "deepseek-official", + "api": "anthropic-messages", + "base_url": "https://api.deepseek.com/anthropic", + "reasoning_effort": "high", + "profile": "sdk-minimal", + "temperature": "unset", + "sdk_max_tokens": "unset", + "upstream_default_max_tokens": 256000, + "alias": "Official guide checked 2026-11-01 maps the mutable deepseek-flash alias to DeepSeek-V4.1-Flash-1017." + }, + "budget_policy": "Original task-native timeouts/resources; 1800-second setup. Upstream sdk-minimal prompts, tools, compaction, retries, and generation defaults. No injected turn limit, internal time budget, or finalization reminders. Pi used OpenAI-compatible transport with 384000-token model metadata; this is not a matched-generation-budget causal harness comparison.", + "verifier_policy": "Original official tests and native per-attempt timeout; retry only verifier timeouts, at most three total attempts against unchanged agent output. No dependency-error recovery. Require confirmed completion or cancellation before verification.", + "preflight_policy": "Fresh install-only checks for all six tasks before any supplemental model job. All six exact revisions have reviewed dependency-only profiles, original package versions/indexes and pytest --version checks. Every formal job repeats setup before model execution. Successful preflight does not guarantee future network availability.", + "information_policy": "Outcome-selected supplemental development experiment explicitly requested after auditing the original run. Do not replace original rewards or claim an independent clean benchmark. No prior outputs, solutions, official assertions or diagnostics are passed to the model. Inspect only operational metadata until all supplemental outputs are frozen. The other 55 tasks remain unopened.", + "runtime_environment": "UV_PROJECT_ENVIRONMENT=/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/dsh-harbor-venv", + "status_at_registration": "No supplemental model executions; dependency preflight pending.", + "historical_pi_results": "benchmarks/harbor/results/pi-deepseek-flash-dev34-20261024.json", + "authorization": { + "kind": "separately authorized fresh supplemental attempts", + "requested_at_date": "2026-11-01", + "scope": "Rerun all six analytically unscored tasks from the first official Harness dev34 run. Keep both attempts. Prior authorization to use the configured official DeepSeek credential remains applicable.", + "original_results": "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json", + "original_results_sha256": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396", + "original_freeze": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json", + "original_freeze_sha256": "cc421c1429450ac41d355d0df8333f4ab10ca4da0bf47f24bc3892d0724d9c24" + }, + "transport_amendment": "Set UV_HTTP_TIMEOUT=120 seconds in the task environment to tolerate dependency-download stalls (previous default 30). Package versions, sources, official scripts and task-native agent/verifier time limits are unchanged. Existing verified APT and uv/Python release caches are reused. No new wheel mirror or unverified cache." +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json new file mode 100644 index 0000000..01ae805 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json @@ -0,0 +1,60 @@ +{ + "observed_at": "2026-10-24T14:07:09.186521+00:00", + "images": [ + { + "task": "terminal-bench/multi-source-data-merger", + "image": "alexgshaw/multi-source-data-merger:20251031", + "image_id": "sha256:8b32782078ff7383a1b4e5d3cecca8ce287e30f50bb4c0e5db009a18064e666e", + "repo_digests": [ + "alexgshaw/multi-source-data-merger@sha256:8b32782078ff7383a1b4e5d3cecca8ce287e30f50bb4c0e5db009a18064e666e" + ], + "size_bytes": 220880128 + }, + { + "task": "terminal-bench/pytorch-model-cli", + "image": "alexgshaw/pytorch-model-cli:20251031", + "image_id": "sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "repo_digests": [ + "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67" + ], + "size_bytes": 44349611 + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "image": "alexgshaw/pytorch-model-recovery:20260430", + "image_id": "sha256:7c2bb52851cef25bdc3939e473496d25fbf11fd0cb7ab58bcf3cefcb751cdc0e", + "repo_digests": [ + "alexgshaw/pytorch-model-recovery@sha256:7c2bb52851cef25bdc3939e473496d25fbf11fd0cb7ab58bcf3cefcb751cdc0e" + ], + "size_bytes": 269661811 + }, + { + "task": "terminal-bench/query-optimize", + "image": "alexgshaw/query-optimize:20251031", + "image_id": "sha256:28de7da87004eab5b29bd5eaf1823579bc5b73ccea4b80349a5623185770da14", + "repo_digests": [ + "alexgshaw/query-optimize@sha256:28de7da87004eab5b29bd5eaf1823579bc5b73ccea4b80349a5623185770da14" + ], + "size_bytes": 120494963 + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "image": "alexgshaw/torch-pipeline-parallelism:20251031", + "image_id": "sha256:3cb7b39d86f1704b20e668851da0b08b34b3270c01d583f136d65190984a8d1b", + "repo_digests": [ + "alexgshaw/torch-pipeline-parallelism@sha256:3cb7b39d86f1704b20e668851da0b08b34b3270c01d583f136d65190984a8d1b" + ], + "size_bytes": 29725531 + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "image": "alexgshaw/torch-tensor-parallelism:20251031", + "image_id": "sha256:c50ac169e11465f41b49749fe87e4487f24d31a0c39c0906cd3242cc5499c690", + "repo_digests": [ + "alexgshaw/torch-tensor-parallelism@sha256:c50ac169e11465f41b49749fe87e4487f24d31a0c39c0906cd3242cc5499c690" + ], + "size_bytes": 29725529 + } + ], + "matches_prior_registration": true +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json new file mode 100644 index 0000000..eb64222 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json new file mode 100644 index 0000000..b7303e5 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + }, + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json new file mode 100644 index 0000000..5aa2fa6 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json new file mode 100644 index 0000000..6bfaade --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json @@ -0,0 +1,9 @@ +{ + "checked_at": "2026-11-01T14:36:34.471637+00:00", + "source": "https://api-docs.deepseek.com/quick_start/agent_integrations/", + "alias": "deepseek-flash", + "documented_serving_model": "DeepSeek-V4.1-Flash-1017", + "local_response": "jobs/deepseek-alias-docs-20261101.html", + "local_response_sha256": "723bbdcf0b8fd45ae629b0c9fa3939bbfa2af368c69c2ce37ff532049ddde9e3", + "limitation": "Mutable alias; documentation does not establish an immutable serving checkpoint across dates." +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json new file mode 100644 index 0000000..22cb9ad --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json new file mode 100644 index 0000000..099daf2 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json new file mode 100644 index 0000000..fe9cf6c --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json new file mode 100644 index 0000000..d22861e --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json new file mode 100644 index 0000000..67e201a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json @@ -0,0 +1,13 @@ +{ + "selection_policy": "All six and only the six analytically unscored original tasks, explicitly requested by the maintainer on 2026-11-01. Fresh supplemental attempts; no best-of selection or replacement of prior records.", + "original_results": "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json", + "original_results_sha256": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396", + "tasks": [ + "terminal-bench/multi-source-data-merger", + "terminal-bench/pytorch-model-cli", + "terminal-bench/pytorch-model-recovery", + "terminal-bench/query-optimize", + "terminal-bench/torch-pipeline-parallelism", + "terminal-bench/torch-tensor-parallelism" + ] +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json new file mode 100644 index 0000000..80e573a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json new file mode 100644 index 0000000..4739479 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} From 80b468bbe34043a14ddb91b25405c99fc6eb008c Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 07:21:27 -0800 Subject: [PATCH 10/15] feat(benchmarks): verify and cache original pypi wheels --- benchmarks/harbor/README.md | 31 +++ benchmarks/harbor/README.zh-CN.md | 25 ++ benchmarks/harbor/scripts/pypi_wheel_cache.py | 234 ++++++++++++++++++ .../harbor/tests/test_pypi_wheel_cache.py | 189 ++++++++++++++ 4 files changed, 479 insertions(+) create mode 100644 benchmarks/harbor/scripts/pypi_wheel_cache.py create mode 100644 benchmarks/harbor/tests/test_pypi_wheel_cache.py diff --git a/benchmarks/harbor/README.md b/benchmarks/harbor/README.md index d34cd27..b30cc90 100644 --- a/benchmarks/harbor/README.md +++ b/benchmarks/harbor/README.md @@ -104,6 +104,37 @@ when it runs; a successful preflight is not a guarantee against later outages. Full standard runs enable the timeout retry policy below by default, with zero whole-trial retries. CI checks these contracts without model calls or Docker. +### Verified PyPI wheel transport + +Large original PyPI wheels can use the optional loopback cache: + +```bash +python benchmarks/harbor/scripts/pypi_wheel_cache.py \ + --cache .cache/nanopy-harbor/pypi-ranges --bind 127.0.0.1 --port 18744 +``` + +For Docker tasks, run it in a sidecar sharing the task's network namespace and +mount the cache only in that sidecar. Register +`UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple` in the task environment. +The service always fetches current PyPI index metadata, preserving all versions, +requirements, yanks, hashes, and extra-index settings. Only wheel URLs change to +the loopback cache. Source-distribution URLs and PEP 658 metadata bytes remain +original; metadata is forwarded fresh. Every wheel request rechecks its +filename, size, SHA256, and original HTTPS URL against current PyPI release +metadata before restoring any cached bytes. Metadata outages and corrupt files +fail closed. The cache is not an offline or alternative package index. + +A cold wheel is fully downloaded and verified before its body is served. +Allow a suitable uv HTTP timeout (for example `UV_HTTP_TIMEOUT=600`) within +the separately registered setup deadline. Native agent and verifier timeouts +remain unchanged. + +The existing bounded range downloader verifies every range and the complete +wheel digest. File locks protect publication across task sidecars. Register the +transport and its scripts before model work, run dependency-only preflight, +and preserve original task images, scripts, package versions, sources, and +per-attempt agent/verifier limits. + ### Pi agent baselines The same entry point also supports `harbor_adapter:PiCodeAgent`. Register a local diff --git a/benchmarks/harbor/README.zh-CN.md b/benchmarks/harbor/README.zh-CN.md index e21030f..5296e74 100644 --- a/benchmarks/harbor/README.zh-CN.md +++ b/benchmarks/harbor/README.zh-CN.md @@ -83,6 +83,31 @@ CPU 索引和 `unsafe-best-match` 策略;其他 PyTorch 配置保留原始默 通过不能保证后续没有网络故障。完整标准运行默认启用下述验证超时重试,整题重试 次数为零。CI 会验证这些约定,不调用模型、不启动 Docker。 +### 校验后的 PyPI wheel 传输 + +较大的原始 PyPI wheel 可使用可选的本地回环缓存: + +```bash +python benchmarks/harbor/scripts/pypi_wheel_cache.py \ + --cache .cache/nanopy-harbor/pypi-ranges --bind 127.0.0.1 --port 18744 +``` + +Docker 题目中,让 sidecar 与题目共享网络命名空间,并仅向 sidecar 挂载缓存。 +在题目环境中注册 `UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple`。 +服务始终获取当前 PyPI 索引元数据,保留所有版本、依赖要求、撤回标记、哈希和 +额外索引设置,仅将 wheel URL 指向回环缓存。源码分发 URL 与 PEP 658 元数据 +内容保持原样,元数据每次从上游转发。每次 wheel 请求都先根据当前 PyPI 发布 +元数据重新核对文件名、大小、SHA256 和原始 HTTPS URL,再恢复缓存内容。 +元数据不可用或文件损坏时直接失败,不使用离线或替代软件源。 + +首次请求的 wheel 会先完整下载并校验,再返回内容。应在单独注册的安装时限 +范围内设置合适的 uv HTTP 等待时间(例如 `UV_HTTP_TIMEOUT=600`)。原生 +agent 和验证器时限保持不变。 + +已有的有界分段下载器会校验每个范围及整个 wheel 的摘要。文件锁保护多个 +题目 sidecar 之间的缓存发布。模型运行前应注册传输配置及脚本,执行仅依赖 +预检,并保留原始题目镜像、脚本、软件包版本、来源以及每次 agent / 验证器时限。 + ### Pi agent baseline 同一个标准入口也支持 `harbor_adapter:PiCodeAgent`。在 agent kwargs 中用 diff --git a/benchmarks/harbor/scripts/pypi_wheel_cache.py b/benchmarks/harbor/scripts/pypi_wheel_cache.py new file mode 100644 index 0000000..01b7e21 --- /dev/null +++ b/benchmarks/harbor/scripts/pypi_wheel_cache.py @@ -0,0 +1,234 @@ +"""Serve the original PyPI index with hash-verified, ranged wheel downloads. + +Point uv's default index at this loopback service. Project metadata is always +fetched from PyPI; versions, requirements, yanks, hashes and extra indexes are +preserved. Only wheel URLs are routed through the cache. Every wheel request +rechecks the current PyPI release metadata before serving any cached bytes. +""" + +import argparse +from datetime import datetime, timezone +import fcntl +from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer +import importlib.util +import json +from pathlib import Path +import re +import threading +from urllib.parse import quote, unquote, urlsplit + + +spec = importlib.util.spec_from_file_location( + "pypi_verified_ranges", Path(__file__).with_name("cache_image.py"), +) +ranges = importlib.util.module_from_spec(spec) +spec.loader.exec_module(ranges) +PROJECT = re.compile(r"[a-z0-9]+(?:-[a-z0-9]+)*") +WHEEL = re.compile(r"[A-Za-z0-9_.+!-]+\.whl") +DIGEST = re.compile(r"[a-f0-9]{64}") + + +def canonical(name): + return re.sub(r"[-_.]+", "-", name).lower() + + +def event(**values): + print(json.dumps({"at": datetime.now(timezone.utc).isoformat(), **values}), flush=True) + + +def public_json(url, *, accept="application/json"): + status, _, body = ranges.get(url, {"Accept": accept}, 32 * 1024 * 1024) + if status != 200: + raise ValueError("PyPI metadata request did not succeed") + return json.loads(body) + + +def official_wheel_url(url, filename): + parsed = urlsplit(url) + if (parsed.scheme != "https" or parsed.netloc != "files.pythonhosted.org" + or parsed.query or parsed.fragment + or unquote(parsed.path.rsplit("/", 1)[-1]) != filename): + raise ValueError("Wheel must retain its original PyPI file URL") + return url + + +class WheelCache: + def __init__(self, root, *, workers=16, downloads=2): + self.root = Path(root) + self.workers = workers + self.slots = threading.BoundedSemaphore(downloads) + (self.root / "locks").mkdir(parents=True, exist_ok=True) + + def index(self, project): + if not PROJECT.fullmatch(project): + raise ValueError("Invalid normalized project name") + data = public_json( + f"https://pypi.org/simple/{project}/", + accept="application/vnd.pypi.simple.v1+json", + ) + for file in data["files"]: + name = file["filename"] + if not WHEEL.fullmatch(name): + continue # Source distributions retain their original URLs. + digest = file["hashes"]["sha256"] + if not DIGEST.fullmatch(digest): + raise ValueError("PyPI wheel lacks a SHA256 digest") + official_wheel_url(file["url"], name) + file["url"] = f"/files/{project}/{digest}/{quote(name, safe='')}" + return data + + def release_file(self, project, digest, filename): + if (not PROJECT.fullmatch(project) or not DIGEST.fullmatch(digest) + or not WHEEL.fullmatch(filename)): + raise ValueError("Invalid wheel request") + parts = filename.split("-") + if len(parts) < 5 or canonical(parts[0]) != project: + raise ValueError("Wheel project differs from its index") + version = parts[1] + data = public_json(f"https://pypi.org/pypi/{project}/{quote(version, safe='')}/json") + file = next((f for f in data["urls"] if f["filename"] == filename), None) + if (file is None or file["digests"]["sha256"] != digest + or type(file["size"]) is not int or file["size"] <= 0): + raise ValueError("Wheel no longer matches current PyPI metadata") + official_wheel_url(file["url"], filename) + return file + + def wheel(self, file): + digest = file["digests"]["sha256"] + descriptor = {"digest": "sha256:" + digest, "size": file["size"]} + # A host bind mount is shared by separate task-sidecar processes too. + # Serialize publication across processes, not merely HTTP threads. + with (self.root / "locks" / digest).open("a") as lock: + fcntl.flock(lock, fcntl.LOCK_EX) + cached = ranges.blob_path(self.root, descriptor).exists() + with self.slots: + path = ranges.fetch_blob( + self.root, descriptor, file["url"], {}, workers=self.workers, + chunk_size=4 * 1024 * 1024, + ) + event(kind="wheel", filename=file["filename"], size=file["size"], + sha256=digest, cache_hit=cached) + return path + + +def byte_range(header, size): + if header is None: + return 0, size - 1, 200 + match = re.fullmatch(r"bytes=(\d*)-(\d*)", header) + if not match or not any(match.groups()): + raise ValueError("Unsupported byte range") + first, last = match.groups() + if first: + start, end = int(first), min(int(last), size - 1) if last else size - 1 + else: + start, end = max(0, size - int(last)), size - 1 + if start >= size or start > end: + raise ValueError("Unsatisfiable byte range") + return start, end, 206 + + +class Handler(BaseHTTPRequestHandler): + protocol_version = "HTTP/1.1" + + def log_message(self, *args): + pass + + def do_GET(self): + self.respond(False) + + def do_HEAD(self): + self.respond(True) + + def headers_for(self, status, size, content_type, **headers): + self.send_response(status) + self.send_header("Content-Length", str(size)) + self.send_header("Content-Type", content_type) + self.send_header("Connection", "close") + for name, value in headers.items(): + self.send_header(name.replace("_", "-"), str(value)) + self.end_headers() + self.response_started = True + + def respond(self, head): + self.close_connection = True + self.response_started = False + path = urlsplit(self.path) + if path.query or path.fragment: + self.send_error(400, "Unexpected query") + return + try: + if path.path == "/health": + self.headers_for(200, 0, "text/plain") + return + index = re.fullmatch(r"/simple/([a-z0-9-]+)/?", path.path) + if index: + body = json.dumps(self.server.cache.index(index[1])).encode() + self.headers_for(200, len(body), "application/vnd.pypi.simple.v1+json", Cache_Control="no-cache") + if not head: + self.wfile.write(body) + return + match = re.fullmatch(r"/files/([^/]+)/([^/]+)/([^/]+)", path.path) + if not match: + self.send_error(404, "Unknown cache endpoint") + return + project, digest, filename = (unquote(value) for value in match.groups()) + metadata = filename.endswith(".whl.metadata") + if metadata: + filename = filename.removesuffix(".metadata") + file = self.server.cache.release_file(project, digest, filename) + if metadata: + # Preserve the original PEP 658 bytes and advertised hashes. + # Metadata is forwarded fresh, never restored from this cache. + status, _, body = ranges.get(file["url"] + ".metadata", limit=4 * 1024 * 1024) + if status != 200: + raise ValueError("Original wheel metadata is unavailable") + self.headers_for(200, len(body), "application/octet-stream", Cache_Control="no-cache") + if not head: + self.wfile.write(body) + return + try: + start, end, status = byte_range(self.headers.get("Range"), file["size"]) + except ValueError: + self.headers_for(416, 0, "application/octet-stream", Content_Range=f"bytes */{file['size']}") + return + blob = None if head else self.server.cache.wheel(file) + headers = {"Accept_Ranges": "bytes", "ETag": '"' + digest + '"'} + if status == 206: + headers["Content_Range"] = f"bytes {start}-{end}/{file['size']}" + self.headers_for(status, end - start + 1, "application/octet-stream", **headers) + if not head: + with blob.open("rb") as stream: + stream.seek(start) + remaining = end - start + 1 + while remaining: + block = stream.read(min(remaining, 64 * 1024)) + if not block: + return # Incomplete transfers cannot fabricate valid wheel bytes. + self.wfile.write(block) + remaining -= len(block) + except (BrokenPipeError, ConnectionResetError): + pass + except Exception as error: + event(kind="error", error_type=type(error).__name__) + if not self.response_started: + self.send_error(502, "Original PyPI metadata or wheel validation failed") + + +def main(): + parser = argparse.ArgumentParser(description=__doc__) + parser.add_argument("--cache", type=Path, required=True) + parser.add_argument("--bind", default="127.0.0.1") + parser.add_argument("--port", type=int, default=18744) + parser.add_argument("--workers", type=int, default=16) + parser.add_argument("--downloads", type=int, default=2) + args = parser.parse_args() + if not 1 <= args.workers <= 32 or not 1 <= args.downloads <= 4: + parser.error("Use 1-32 workers and 1-4 concurrent wheel downloads") + server = ThreadingHTTPServer((args.bind, args.port), Handler) + server.cache = WheelCache(args.cache, workers=args.workers, downloads=args.downloads) + event(kind="ready", bind=args.bind, port=args.port) + server.serve_forever() + + +if __name__ == "__main__": + main() diff --git a/benchmarks/harbor/tests/test_pypi_wheel_cache.py b/benchmarks/harbor/tests/test_pypi_wheel_cache.py new file mode 100644 index 0000000..d734d0b --- /dev/null +++ b/benchmarks/harbor/tests/test_pypi_wheel_cache.py @@ -0,0 +1,189 @@ +"""Original-index preservation and fail-closed wheel cache contracts.""" + +from concurrent.futures import ThreadPoolExecutor +import copy +import hashlib +from http.server import ThreadingHTTPServer +import importlib.util +import json +from pathlib import Path +import threading +import time +from urllib.error import HTTPError +from urllib.request import Request, urlopen + +import pytest + + +spec = importlib.util.spec_from_file_location( + "pypi_cache_test", Path(__file__).parents[1] / "scripts/pypi_wheel_cache.py", +) +proxy = importlib.util.module_from_spec(spec) +spec.loader.exec_module(proxy) + + +@pytest.fixture +def origin(tmp_path, monkeypatch): + content = b"original wheel bytes" + metadata = b"Metadata-Version: 2.1\nName: demo\nVersion: 1.0\n" + digest = hashlib.sha256(content).hexdigest() + name = "demo-1.0-py3-none-any.whl" + url = "https://files.pythonhosted.org/packages/original/" + name + file = {"filename": name, "size": len(content), "url": url, + "digests": {"sha256": digest}} + release = {"urls": [file]} + index = {"meta": {"api-version": "1.4"}, "name": "demo", "files": [{ + "filename": name, "size": len(content), "url": url, + "hashes": {"sha256": digest}, "requires-python": ">=3.10", "yanked": "test yank reason", + "core-metadata": {"sha256": hashlib.sha256(metadata).hexdigest()}, + }, {"filename": "demo-1.0.tar.gz", "url": "https://files.pythonhosted.org/demo-1.0.tar.gz"}]} + requests = [] + + def public_json(url, **kwargs): + requests.append(url) + if url == "https://pypi.org/simple/demo/": + return copy.deepcopy(index) + assert url == "https://pypi.org/pypi/demo/1.0/json" + return copy.deepcopy(release) + + def get(request_url, headers=None, limit=None): + requests.append(request_url) + if request_url == url + ".metadata": + return 200, {}, metadata + assert request_url == url + assert headers["Range"] == f"bytes=0-{len(content) - 1}" + return 206, {"Content-Range": f"bytes 0-{len(content) - 1}/{len(content)}"}, content + + monkeypatch.setattr(proxy, "public_json", public_json) + monkeypatch.setattr(proxy.ranges, "get", get) + cache = proxy.WheelCache(tmp_path / "cache") + return cache, file, release, index, content, metadata, requests + + +def test_index_changes_only_wheel_urls_and_refreshes_origin(origin): + cache, file, _, index, _, _, requests = origin + actual = cache.index("demo") + expected = copy.deepcopy(index) + expected["files"][0]["url"] = f"/files/demo/{file['digests']['sha256']}/{file['filename']}" + assert actual == expected + index["files"][0]["yanked"] = False + assert cache.index("demo")["files"][0]["yanked"] is False + assert requests.count("https://pypi.org/simple/demo/") == 2 + + +@pytest.mark.parametrize("change", ["filename", "size", "sha256", "url", "missing"]) +def test_cached_wheel_is_rejected_when_current_metadata_changes(origin, change): + cache, file, release, _, content, _, requests = origin + name, digest = file["filename"], file["digests"]["sha256"] + assert cache.wheel(cache.release_file("demo", digest, name)).read_bytes() == content + if change == "missing": + release["urls"] = [] + elif change == "sha256": + file["digests"]["sha256"] = "0" * 64 + elif change == "size": + file["size"] += 1 + elif change == "url": + file["url"] = "https://example.com/" + name + else: + file["filename"] = "demo-1.1-py3-none-any.whl" + with pytest.raises(ValueError): + cache.wheel(cache.release_file("demo", digest, name)) + assert requests.count("https://pypi.org/pypi/demo/1.0/json") == 2 + + +def test_corrupt_cached_bytes_fail_closed_before_serving(origin): + cache, file, _, _, content, _, _ = origin + path = cache.wheel(file) + path.write_bytes(b"x" * len(content)) + with pytest.raises(ValueError, match="SHA256"): + cache.wheel(cache.release_file("demo", file["digests"]["sha256"], file["filename"])) + + +@pytest.mark.parametrize("project,digest,name", [ + ("../demo", "a" * 64, "demo-1.0-py3-none-any.whl"), + ("demo", "../escape", "demo-1.0-py3-none-any.whl"), + ("demo", "a" * 64, "../demo-1.0-py3-none-any.whl"), + ("other", "a" * 64, "demo-1.0-py3-none-any.whl"), +]) +def test_request_paths_cannot_escape_or_change_the_project(origin, project, digest, name): + cache, _, _, _, _, _, requests = origin + with pytest.raises(ValueError): + cache.release_file(project, digest, name) + assert not requests + + +def test_separate_cache_instances_serialize_shared_blob_publication(origin, monkeypatch): + cache, file, _, _, _, _, _ = origin + other = proxy.WheelCache(cache.root) + active = 0 + maximum = 0 + guard = threading.Lock() + + def fetch(*args, **kwargs): + nonlocal active, maximum + with guard: + active += 1 + maximum = max(maximum, active) + time.sleep(0.03) + with guard: + active -= 1 + return cache.root / "placeholder" + + monkeypatch.setattr(proxy.ranges, "fetch_blob", fetch) + with ThreadPoolExecutor(max_workers=2) as pool: + list(pool.map(lambda item: item.wheel(file), [cache, other])) + assert maximum == 1 + + +@pytest.fixture +def server(origin): + cache = origin[0] + server = ThreadingHTTPServer(("127.0.0.1", 0), proxy.Handler) + server.cache = cache + thread = threading.Thread(target=server.serve_forever, daemon=True) + thread.start() + try: + yield f"http://127.0.0.1:{server.server_port}" + finally: + server.shutdown() + server.server_close() + thread.join() + + +def wheel_url(server, origin): + file = origin[1] + return server + f"/files/demo/{file['digests']['sha256']}/{file['filename']}" + + +def test_http_index_head_and_range_deliver_original_bytes(server, origin): + _, file, _, _, content, _, _ = origin + with urlopen(server + "/simple/demo/") as response: + assert json.load(response)["files"][0]["url"].endswith(file["filename"]) + assert response.headers["Cache-Control"] == "no-cache" + url = wheel_url(server, origin) + with urlopen(Request(url, method="HEAD")) as response: + assert int(response.headers["Content-Length"]) == len(content) + assert response.read() == b"" + for requested, expected in [("bytes=2-6", content[2:7]), ("bytes=-5", content[-5:])]: + with urlopen(Request(url, headers={"Range": requested})) as response: + assert response.status == 206 + assert response.read() == expected + with pytest.raises(HTTPError) as error: + urlopen(Request(url, headers={"Range": "bytes=999-"})) + assert error.value.code == 416 + + +def test_http_metadata_is_forwarded_fresh_and_keeps_original_bytes(server, origin): + _, file, _, _, _, metadata, requests = origin + for _ in range(2): + with urlopen(wheel_url(server, origin) + ".metadata") as response: + assert response.read() == metadata + assert requests.count(file["url"] + ".metadata") == 2 + + +def test_http_corrupt_cache_never_returns_successful_content(server, origin): + cache, file, _, _, content, _, _ = origin + cache.wheel(file).write_bytes(b"x" * len(content)) + with pytest.raises(HTTPError) as error: + urlopen(wheel_url(server, origin)) + assert error.value.code == 502 From dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 07:27:08 -0800 Subject: [PATCH 11/15] feat(benchmarks): retain verified pypi cache provenance --- benchmarks/harbor/README.md | 4 ++++ benchmarks/harbor/README.zh-CN.md | 3 +++ benchmarks/harbor/scripts/pypi_wheel_cache.py | 20 ++++++++++++++++++- .../harbor/tests/test_pypi_wheel_cache.py | 13 ++++++++++++ 4 files changed, 39 insertions(+), 1 deletion(-) diff --git a/benchmarks/harbor/README.md b/benchmarks/harbor/README.md index b30cc90..5191fd1 100644 --- a/benchmarks/harbor/README.md +++ b/benchmarks/harbor/README.md @@ -135,6 +135,10 @@ transport and its scripts before model work, run dependency-only preflight, and preserve original task images, scripts, package versions, sources, and per-attempt agent/verifier limits. +Use `--audit-log /cache/run-events.jsonl` to retain current release metadata +and verified cache-hit events after the sidecars exit. The append-only log is +locked across processes and includes the originating container instance. + ### Pi agent baselines The same entry point also supports `harbor_adapter:PiCodeAgent`. Register a local diff --git a/benchmarks/harbor/README.zh-CN.md b/benchmarks/harbor/README.zh-CN.md index 5296e74..3ca5d75 100644 --- a/benchmarks/harbor/README.zh-CN.md +++ b/benchmarks/harbor/README.zh-CN.md @@ -108,6 +108,9 @@ agent 和验证器时限保持不变。 题目 sidecar 之间的缓存发布。模型运行前应注册传输配置及脚本,执行仅依赖 预检,并保留原始题目镜像、脚本、软件包版本、来源以及每次 agent / 验证器时限。 +使用 `--audit-log /cache/run-events.jsonl`,可在 sidecar 退出后保留当前发布 +元数据及已校验的缓存命中事件。追加日志使用跨进程锁,并记录来源容器实例。 + ### Pi agent baseline 同一个标准入口也支持 `harbor_adapter:PiCodeAgent`。在 agent kwargs 中用 diff --git a/benchmarks/harbor/scripts/pypi_wheel_cache.py b/benchmarks/harbor/scripts/pypi_wheel_cache.py index 01b7e21..eca223e 100644 --- a/benchmarks/harbor/scripts/pypi_wheel_cache.py +++ b/benchmarks/harbor/scripts/pypi_wheel_cache.py @@ -12,6 +12,7 @@ from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer import importlib.util import json +import os from pathlib import Path import re import threading @@ -26,6 +27,7 @@ PROJECT = re.compile(r"[a-z0-9]+(?:-[a-z0-9]+)*") WHEEL = re.compile(r"[A-Za-z0-9_.+!-]+\.whl") DIGEST = re.compile(r"[a-f0-9]{64}") +AUDIT_LOG = None def canonical(name): @@ -33,7 +35,14 @@ def canonical(name): def event(**values): - print(json.dumps({"at": datetime.now(timezone.utc).isoformat(), **values}), flush=True) + line = json.dumps({"at": datetime.now(timezone.utc).isoformat(), + "instance": os.environ.get("HOSTNAME", "local"), **values}) + if AUDIT_LOG is not None: + with AUDIT_LOG.open("a") as stream: + fcntl.flock(stream, fcntl.LOCK_EX) + stream.write(line + "\n") + stream.flush() + print(line, flush=True) def public_json(url, *, accept="application/json"): @@ -91,6 +100,9 @@ def release_file(self, project, digest, filename): or type(file["size"]) is not int or file["size"] <= 0): raise ValueError("Wheel no longer matches current PyPI metadata") official_wheel_url(file["url"], filename) + event(kind="release_metadata", project=project, version=version, + filename=filename, size=file["size"], sha256=digest, + source_url=file["url"]) return file def wheel(self, file): @@ -215,15 +227,21 @@ def respond(self, head): def main(): + global AUDIT_LOG parser = argparse.ArgumentParser(description=__doc__) parser.add_argument("--cache", type=Path, required=True) parser.add_argument("--bind", default="127.0.0.1") parser.add_argument("--port", type=int, default=18744) parser.add_argument("--workers", type=int, default=16) parser.add_argument("--downloads", type=int, default=2) + parser.add_argument("--audit-log", type=Path, + help="Append public metadata and verified cache events for the run") args = parser.parse_args() if not 1 <= args.workers <= 32 or not 1 <= args.downloads <= 4: parser.error("Use 1-32 workers and 1-4 concurrent wheel downloads") + AUDIT_LOG = args.audit_log + if AUDIT_LOG is not None: + AUDIT_LOG.parent.mkdir(parents=True, exist_ok=True) server = ThreadingHTTPServer((args.bind, args.port), Handler) server.cache = WheelCache(args.cache, workers=args.workers, downloads=args.downloads) event(kind="ready", bind=args.bind, port=args.port) diff --git a/benchmarks/harbor/tests/test_pypi_wheel_cache.py b/benchmarks/harbor/tests/test_pypi_wheel_cache.py index d734d0b..7bf305f 100644 --- a/benchmarks/harbor/tests/test_pypi_wheel_cache.py +++ b/benchmarks/harbor/tests/test_pypi_wheel_cache.py @@ -99,6 +99,19 @@ def test_corrupt_cached_bytes_fail_closed_before_serving(origin): cache.wheel(cache.release_file("demo", file["digests"]["sha256"], file["filename"])) +def test_audit_retains_current_origin_identity_and_verified_cache_hit(origin, tmp_path, monkeypatch): + cache, file, _, _, _, _, _ = origin + log = tmp_path / "audit.jsonl" + monkeypatch.setattr(proxy, "AUDIT_LOG", log) + for _ in range(2): + cache.wheel(cache.release_file("demo", file["digests"]["sha256"], file["filename"])) + events = [json.loads(line) for line in log.read_text().splitlines()] + assert [event["kind"] for event in events] == ["release_metadata", "wheel"] * 2 + assert events[0]["source_url"] == file["url"] + assert all(event["sha256"] == file["digests"]["sha256"] for event in events) + assert [events[i]["cache_hit"] for i in (1, 3)] == [False, True] + + @pytest.mark.parametrize("project,digest,name", [ ("../demo", "a" * 64, "demo-1.0-py3-none-any.whl"), ("demo", "../escape", "demo-1.0-py3-none-any.whl"), From 0c587c016fc0757f08b223a1976fae9e800d352b Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 07:30:13 -0800 Subject: [PATCH 12/15] test(benchmarks): register verified wheel transport for the supplement --- .../README.md | 39 +- .../experiment-v2.json | 345 ++++++++++++++++++ .../preflight-v1-evidence.json | 136 +++++++ .../source-comparison.json | 17 + .../dependency-cache.compose.json | 148 ++++++++ .../transport-v2/install-pair-01.json | 42 +++ .../transport-v2/install-pair-02.json | 42 +++ .../transport-v2/install-pair-03.json | 42 +++ .../multi-source-data-merger.json | 38 ++ .../transport-v2/pytorch-model-cli.json | 38 ++ .../transport-v2/pytorch-model-recovery.json | 38 ++ .../transport-v2/query-optimize.json | 38 ++ .../torch-pipeline-parallelism.json | 38 ++ .../torch-tensor-parallelism.json | 38 ++ .../transport-v2/transport-probes.json | 71 ++++ 15 files changed, 1100 insertions(+), 10 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/source-comparison.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md index 3d1b069..803416b 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md @@ -1,7 +1,16 @@ # Authorized six-task DeepSeek Harness supplement -Status: registered on November 1, 2026; dependency preflight pending. No -supplemental model executions have started. +Status: the [v2 registration](experiment-v2.json) adds verified PyPI wheel +transport after the initial preparation encountered a setup timeout. Fresh v2 +dependency preflight is pending; no supplemental model executions have started. + +The initial direct-download preflight passed data merger, PyTorch CLI, and query +optimization. Model recovery exceeded its 1,800-second setup limit while +preparing verifier dependencies. The controller stopped after that completed +second pair, before the third pair or any model job; no active trial was +interrupted. [Initial evidence](preflight-v1-evidence.json) preserves all four +trials. The original registration remains immutable; its model queue is +superseded, not executed. The maintainer explicitly requested another run of the six tasks left analytically unscored in the [original dev34 results](../../results/deepseek-harness-dev34-20261030.json). @@ -31,12 +40,22 @@ it adds reviewed dependency-only preflight profiles for the five previously unsupported tasks in this subset. Model recovery retains its existing profile. The commands preserve the original Python version, package pins, APT requirements, and index options, and run only `pytest --version` from `/tmp`. -All 285 Harbor tests passed, including regressions for those profiles. +All 285 Harbor tests passed for the dependency-profile addition. The final +transport implementation passed all 301 Harbor tests, including 16 cache +contracts, and a real uv 0.9.5 smoke with a fresh temporary uv cache and zero +model calls. See [transport probes](transport-v2/transport-probes.json). -The task environment sets `UV_HTTP_TIMEOUT=120` seconds instead of the default -30 to tolerate download stalls. It retains original package sources and the -existing verified APT-index and uv/Python-release caches. This transport change -does not increase the task-native agent/verifier limits. Setup remains 1,800 +The initial plan set `UV_HTTP_TIMEOUT=120` instead of the default 30. V2 sets +`UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple` and `UV_HTTP_TIMEOUT=600` to allow +a cold wheel to be fully verified before response bytes are served. The loopback +sidecar forwards current original PyPI metadata, preserving versions, hashes, +requirements, yanks, and extra-index settings. Original wheel bytes are checked +against current release filename, size, SHA256, and HTTPS origin on every +request; cache corruption or unavailable metadata fails closed. Cross-process +locks protect shared downloads. Public metadata/cache events persist in +`.cache/nanopy-harbor/pypi-ranges/audit/dsh-sup6-cache-20261101.jsonl`. +The existing verified APT-index and uv/Python-release caches remain identical. +These transport changes do not increase task-native agent/verifier limits. Setup remains 1,800 seconds; concurrency remains two with pair barriers. There is one fresh model attempt per selected task and no automatic whole-trial retry. Only verifier timeouts may retry, at most three total attempts on the same new output. @@ -53,17 +72,17 @@ export UV_PROJECT_ENVIRONMENT="$PWD/jobs/dsh-harbor-venv" export UV_CACHE_DIR="$PWD/.cache/uv" uv run --frozen --project benchmarks/harbor \ python benchmarks/harbor/scripts/run_comparison.py \ - --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json \ --phase preflight # Supply the previously authorized DEEPSEEK_API_KEY in the environment. uv run --frozen --project benchmarks/harbor \ python benchmarks/harbor/scripts/run_comparison.py \ - --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json \ --phase model ``` Every job uses the standard `python -m harbor_adapter run --config ...` entry point from the shared frozen source. Native outputs remain under `jobs/deepseek-harness-supplement-source/jobs/`; orchestration and image events -remain under `jobs/deepseek-harness-dev6-supplement-20261101-orchestration/`. +remain under `jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/`. The existing original registration and frozen outputs are not overwritten. diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json new file mode 100644 index 0000000..e54ce6e --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json @@ -0,0 +1,345 @@ +{ + "schema_version": 1, + "registered_at": "2026-11-01T15:28:42.169929+00:00", + "experiment": "deepseek-harness-dev6-supplement-20261101-v2", + "total_model_runs": 6, + "max_concurrent_trials": 2, + "whole_trial_retries": 0, + "minimum_free_disk_bytes": 16106127360, + "log_dir": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration", + "arms": { + "deepseek": { + "agent": "deepseek-harness", + "source_root": "jobs/deepseek-harness-supplement-source", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6" + } + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/multi-source-data-merger:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-processing", + "verifier_dependency_profile": "multi-source-data-merger-v1", + "task_file_sha256": { + "README.md": "9bf37162a4c1b88ed93627830bb8512ebae154325d7cb069c883144a0e4c69d6", + "environment/Dockerfile": "c95a40d3d35b473dc15ae1e8784ba7fb1725805c0583fc574dd2764325d75e4f", + "environment/data/source_a/users.json": "75e51e2d3ab0bdba2269d7cf3757576f84f3ee2c13407d2db94289a5958a1778", + "environment/data/source_b/users.csv": "8e87d16dc47d166edd8e56165955bad265b630d53d02762367c816affb5fa760", + "environment/data/source_c/users.parquet": "23d1ea51c4170bdba53febb885a0f495e28de525a52667347b2932136b9a72fe", + "instruction.md": "d1d28bb6e37f7a255a1b7f78aae8591f35c0494f33e4913392424ad954aab944", + "solution/solve.sh": "1b3d0d50e31930d042495d48f4c8e95f1a9c1a979c803b7f7cec651322537e1e", + "task.toml": "ee3964c0ee040f0a0cb121cefefbc15443d127684acb003fa5b61dd42f31c694", + "tests/test.sh": "d5a00e64413c0e1ae452f38a1aa1a47c3ed7b4416bbfccae1a4c053025a336d1", + "tests/test_outputs.py": "a070888930cf829aeca039771260b1386bda990ea805f313468d381ba3608ca5" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pytorch-model-cli:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "model-training", + "verifier_dependency_profile": "pytorch-model-cli-v1", + "task_file_sha256": { + "README.md": "35a093fa5f1592737784ba718c2e3e9ae17ff8b5e53e003fae560ddce091039c", + "environment/Dockerfile": "8fd5c0f21a6914cd440d10c84c9db06fec606f20dc3e0363b17a3b8c220dc4d9", + "environment/task-deps/cJSON.c": "0e7995261640c96db16ed2c91a5d981efa917cf5e72d0f97ec45a1179ae3ad73", + "environment/task-deps/cJSON.h": "0909160541bcc510e0b2d8c7a9cc051fa1459a0f13be4d8e56346a15491f2f2f", + "environment/task-deps/image.png": "f4d5dfefe463939996541404340e7099d5957b56e1df3760d13b5d2de754db53", + "environment/task-deps/lodepng.cpp": "67873c5f6272f779a143a9c81070e8783327c5552d0b9de2aaee2cb779308917", + "environment/task-deps/lodepng.h": "b735b7e0357319973e9488d520d7f270d734a6e9e1bf924fb1d5b8d26d7e97ce", + "environment/task-deps/model.pth": "a32103eef09936e586b1a7dc7e64d279500a38ff30fdd6df2167e28eb954eee6", + "environment/task-deps/model.py": "e4d3a5c728449656d0efde7871ee227619348352769795ed1be1a8b7f835fddc", + "instruction.md": "ce36ee1f6c589f81ee6a6be4967e25c76a2ed697006928023f42dfb9f28a6750", + "solution/solve.sh": "3b34e561c9ef0a53c9948655847846eb4811aaac67b990480aa6aac1b04bfe66", + "task.toml": "fc1c018efbd5d0537ac0741d53d1ebad3886faa51d14cbde8ec953856411bc0d", + "tests/test.sh": "43fb67e537c7adc2df11c2367f47c876cb5d16a7c1ae16d0339c3a3364bb2b06", + "tests/test_outputs.py": "370702c57a4e0c85f70ee37a5c357ae3a5df8480a275bb6263b0376f2f6c093b" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/pytorch-model-recovery:20260430", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "model-training", + "verifier_dependency_profile": "pytorch-model-recovery-v1", + "task_file_sha256": { + "README.md": "e080308a05703b40edd16f53aea41a0e6b9874750ecee35bb432029d0d44bf53", + "environment/Dockerfile": "8a6b7edbb9e7c241e447f15a71e13e5d043b25169be2fd9802b84f5be3d56eb6", + "environment/dataset.pt": "6db65e4ec1c0760209c2dd997ef582e4a1519ab38e0312339999c460a696960d", + "environment/weights.pt": "71471e71c5a311a897362dd5b77413e7ca2ca59abd00a110bdc2cd3f8584d080", + "instruction.md": "2bef07dc28be3ab2c35ae389584034c2e4dde92c2c31b3eb6bd829628c8c3f4a", + "solution/solve.sh": "175a7802777b43e0fa4f629d71f7c68fb40706813c342aaddc72108d89bd5b2c", + "task.toml": "851a51204f53482a31f47cdf8833128cf421113b0331011286940fc857c37dfd", + "tests/test.sh": "6005f3a75ccfe00d7bed1005264b4a84a1f8e30cd780f8a97b3a804be7cc7a8e", + "tests/test_outputs.py": "5d4d57c43e6913bf3d0081da764ad36d1b7275e3bb114a05abda529afe563905", + "tests/weights_gtruth.pt": "71471e71c5a311a897362dd5b77413e7ca2ca59abd00a110bdc2cd3f8584d080" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 1800.0, + "docker_image": "alexgshaw/query-optimize:20251031", + "resources": { + "cpus": 1, + "memory_mb": 2048, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "data-science", + "verifier_dependency_profile": "query-optimize-v1", + "task_file_sha256": { + "README.md": "a82c53e333a0ac0b0b410b5e26cb8e7305dece31db84d340248ffb13d5a45160", + "environment/Dockerfile": "064818771f200e0517d22068f87229a7a95d0a82b0778c84ca7203bf1926a0b8", + "environment/task-deps/my-sql-query.sql": "43382792f0f438625d643e34abe144febd5669518f4ab1ab550a15c6d9314a21", + "instruction.md": "473e12d320d2732f679fb631c2a4dab4a01deea6775686718ae0c417b8a39447", + "solution/solve.sh": "0d95f3cedb287ff9d47c796cbaf812c300d1222c10fa80c49e8acb0e81eb6978", + "task.toml": "6e88f2a86df699594bcb1bf81df4ed694505f79c9420f671335b02524bbe3868", + "tests/golden.sql": "4da1885d56b3ffa7b7b22b422996d018795611f6963499d551582881b74b84fd", + "tests/test.sh": "4770437ea96c3cc84684b4f99d55fb148fcac09f9ea1e8ef49de487716e6c334", + "tests/test_outputs.py": "2068ae80c57af8d52a7ae510f0666732c057806e556b480970f36f14bae683f6" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/torch-pipeline-parallelism:20251031", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "torch-pipeline-parallelism-v1", + "task_file_sha256": { + "README.md": "297ddea07c230dfd246a8ab07337dd7abcd8e5e2a2b2e26590cb0a8af779839c", + "environment/Dockerfile": "c357ea41e4a4013c5c5980b8fef04af6a5ae250ccc6697cf3bad3127401a931f", + "instruction.md": "989bfcf0c54baf6c8cdc391743dbb2f9e858c5779dd85112536875d4d1b2dcd2", + "solution/solve.sh": "787198e4178d88b887697aefdd29819f29435ffdfe356545d92f58a30fb38b70", + "task.toml": "8b4f9f525c08afbbc4eca9168e46238a61ccb903486773e9ca4da64bb7ffe857", + "tests/test.sh": "cc9fd28b6590993fc37e4a99da771cbf04726c8c62fe308e802612241da69dea", + "tests/test_outputs.py": "23574246a407163a8e6581c84287cc2f1eb745efbaad2a317a5a28c39958c64f" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "native_agent_timeout_seconds": 900, + "verifier_timeout_seconds": 900.0, + "docker_image": "alexgshaw/torch-tensor-parallelism:20251031", + "resources": { + "cpus": 1, + "memory_mb": 8192, + "storage_mb": 10240, + "gpus": 0 + }, + "category": "software-engineering", + "verifier_dependency_profile": "torch-tensor-parallelism-v1", + "task_file_sha256": { + "README.md": "4c1b3c3be30548f1f94c6f5503b1bc74b006af7f192dc4cd026f411df1a54604", + "environment/Dockerfile": "5f264ff0bf62a49aee3c6398447de5108df141e7a2ad35c6de8ef39fa62049c6", + "instruction.md": "13bde2ab165d72a311e2bf066d5d6c3c8761bf9d3dd39d1b23eda23bc546b910", + "solution/solve.sh": "d083b7cbf9dbb6cc2ff38c8f2acc150f0059c04aea42d6c2183aeaecc5a19385", + "task.toml": "72fd2a5500d1fe8795dcae76cccb1a29fb25894881a7475f2fa519c87d425a8d", + "tests/test.sh": "a7e6c9408b2401e4e7448371f7ddc72f3284caccfb56277d3cd4e98051868cdf", + "tests/test_outputs.py": "4ae4285e15fba1fcf13d7b7c0638ac8c2e2b0797c937abab1fb6b136fd508290" + }, + "prior_nanopy_work_budget_seconds": 720, + "pi_agent_timeout_seconds": 900 + } + ], + "preflight": [ + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json", + "config_sha256": "a2cdbfd25baa068e6f904fc6843a4657991d5a6c8fb8a7b16ee18ffeba58e9dd", + "job_name": "dsh-sup6-install-pair-01-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json", + "config_sha256": "8260804d67294d30901161efed81d18d6263ab5669e124ae3ea495efd05dbee0", + "job_name": "dsh-sup6-install-pair-02-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json", + "config_sha256": "a2c56775c102aa0acce248d3ca5bf65c70e2c94b3b2805fe738b37e3c461d798", + "job_name": "dsh-sup6-install-pair-03-cache-20261101" + } + ], + "queue": [ + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json", + "config_sha256": "fb57f5183e4f4922815881ba050583175238cc14cf5eb766e7ade1e53f1aaf24", + "task": "terminal-bench/multi-source-data-merger", + "repeat": 1, + "job_name": "dsh-sup6-multi-source-data-merger-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json", + "config_sha256": "ab45588a2c978419e38ebd5aec8cd2b80a8946393e2931930a145554c743b226", + "task": "terminal-bench/pytorch-model-cli", + "repeat": 1, + "job_name": "dsh-sup6-pytorch-model-cli-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json", + "config_sha256": "efb3470630dd176e013bdf739354e23fb640bdd086b9ce17e67d79d8fa71efa5", + "task": "terminal-bench/pytorch-model-recovery", + "repeat": 1, + "job_name": "dsh-sup6-pytorch-model-recovery-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json", + "config_sha256": "253c2b9bd43cdc526cddb1116f4d3a2b27bcd277c73ad0a794fe744120fd841b", + "task": "terminal-bench/query-optimize", + "repeat": 1, + "job_name": "dsh-sup6-query-optimize-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json", + "config_sha256": "d1228591c9e282dbf36effdb3589efed0ea711007be9d7a09d35025a35f54131", + "task": "terminal-bench/torch-pipeline-parallelism", + "repeat": 1, + "job_name": "dsh-sup6-torch-pipeline-parallelism-cache-20261101" + }, + { + "arm": "deepseek", + "config": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json", + "config_sha256": "df3772684d82ca3a55a682f855f28679b8f4d12d46e182bdfdb0e2cb50f66a61", + "task": "terminal-bench/torch-tensor-parallelism", + "repeat": 1, + "job_name": "dsh-sup6-torch-tensor-parallelism-cache-20261101" + } + ], + "additional_file_sha256": { + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf": "d0e1ffb6dbd02c54978ba2df386326311a59f6d3774990180f22bcc6adfaedbc", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json": "34410bdce15ec3b417105eb7f9f6e01a14c47f1bfd0ffe16d76338e7efa51dd2", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json": "560bc83f2a88bb907662db1b60ab15f49443a1572822fa633440f3cc76510557", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json": "77dfda40e45b870a8ce7f07a1e058d416d94a04ab3d27c2f21bc370f22151394", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json": "c2405e16c2aaf2d837b6f6594875e197fd13eb94ed094253e4513c315783d030", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json": "cc4f07cff8a7ec9ffe4de4bdb69322b0ec89b62c72072020d1f9b1f7f0e20e26", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json": "8c918a0998feb41ac3a974228d1bc3b4ba220fb5d76283c77590755738c1d149", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json": "b04b132fa453bec7c4f888e6e52dab533b8870dc1c0ea21881fb887c84ac00ba", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json": "452c5114c9ac5fc82634ce25ce8a2ee0a8f4e3ee15724f5324bdb1befcf865c3", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json": "4e25ecef8ce9beb5a0adfb9977242fe32a9a5559b8f691fa686971d4a6d4e4e1", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json": "a0805861bc826d29e2bc9205c2c720f725f52a549d970c61b407f786dd73df49", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json": "fddffa4e9c5bac957f87460cdd04276f964fb4c2aa1bcd98124387d95f140c99", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json": "2320f05a860f4c91d1fd1a200740cda5ac8fa67ea65dfab3dc3a5b0ef745b6e0", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json": "5047e6d007ed45ee6887bcf09cceb3475b7075e56551a43143c38b8e00991284", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json": "bfa979780e67235b1f5472d025ca7b8cb2ea2b4ef2c7f1fab29d5b4fabee8d5c", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json": "11b91bacfe4d1b2fdcf57a23376ae61b16df9c69802bd5465557db95b95976d8", + "benchmarks/harbor/scripts/run_comparison.py": "228dbff33c0ec8a2359e612da632ef543dcac4273db2f9a7d0573a24e856fbae", + "benchmarks/harbor/scripts/apt_index_proxy.py": "bc15f7adf1d7968d627ceca4683ab86742a7188a1b1955ba1fef37ba48e8a936", + "benchmarks/harbor/scripts/cache_image.py": "11ddde3fe86a6e4bfdefe7c3578794df98d237f701e40f219d76a541911c9f0c", + "benchmarks/harbor/scripts/uv_release_cache.py": "993cc35b1f3a9e8e171b662ab6d4524c5fe059f0d1f0e275a5bb1930111e0ff5", + "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json": "df3772684d82ca3a55a682f855f28679b8f4d12d46e182bdfdb0e2cb50f66a61", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json": "efb3470630dd176e013bdf739354e23fb640bdd086b9ce17e67d79d8fa71efa5", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json": "d734625b2ba1e5624a31ab5164c3bbe8c1b78c60377a366ebe4a9cf4f239569f", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json": "8260804d67294d30901161efed81d18d6263ab5669e124ae3ea495efd05dbee0", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json": "fb57f5183e4f4922815881ba050583175238cc14cf5eb766e7ade1e53f1aaf24", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json": "a2c56775c102aa0acce248d3ca5bf65c70e2c94b3b2805fe738b37e3c461d798", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json": "ab45588a2c978419e38ebd5aec8cd2b80a8946393e2931930a145554c743b226", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json": "d1228591c9e282dbf36effdb3589efed0ea711007be9d7a09d35025a35f54131", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json": "a2cdbfd25baa068e6f904fc6843a4657991d5a6c8fb8a7b16ee18ffeba58e9dd", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json": "b057be706c6139af66a7a460e8853ba83138986a2424b957753dc832e2f265a8", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json": "253c2b9bd43cdc526cddb1116f4d3a2b27bcd277c73ad0a794fe744120fd841b", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json": "cb45a79c8c0b128d6807118a656afc32357de40c250bb6064b9d46798a4232ac", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json": "ead4b47154c9507326f6f03eeada33a2eb486ab027317fc3b8d655df7bd4a7c9", + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/source-comparison.json": "2a9136e82334bea1c4d7a1097537d36227d3e63836a4fc9983709d288e4f749c", + "benchmarks/harbor/scripts/pypi_wheel_cache.py": "ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74" + }, + "batch_images": { + "manifest": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json", + "storage_path": "/var/lib/docker", + "minimum_free_disk_bytes": 16106127360, + "cache_dir": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/images" + }, + "model_policy": { + "model": "deepseek/deepseek-flash", + "provider": "deepseek-official", + "api": "anthropic-messages", + "base_url": "https://api.deepseek.com/anthropic", + "reasoning_effort": "high", + "profile": "sdk-minimal", + "temperature": "unset", + "sdk_max_tokens": "unset", + "upstream_default_max_tokens": 256000, + "alias": "Official guide checked 2026-11-01 maps the mutable deepseek-flash alias to DeepSeek-V4.1-Flash-1017." + }, + "budget_policy": "Original task-native timeouts/resources; 1800-second setup. Upstream sdk-minimal prompts, tools, compaction, retries, and generation defaults. No injected turn limit, internal time budget, or finalization reminders. Pi used OpenAI-compatible transport with 384000-token model metadata; this is not a matched-generation-budget causal harness comparison.", + "verifier_policy": "Original official tests and native per-attempt timeout; retry only verifier timeouts, at most three total attempts against unchanged agent output. No dependency-error recovery. Require confirmed completion or cancellation before verification.", + "preflight_policy": "Fresh install-only checks for all six tasks before any supplemental model job. All six exact revisions have reviewed dependency-only profiles, original package versions/indexes and pytest --version checks. Every formal job repeats setup before model execution. Successful preflight does not guarantee future network availability.", + "information_policy": "Outcome-selected supplemental development experiment explicitly requested after auditing the original run. Do not replace original rewards or claim an independent clean benchmark. No prior outputs, solutions, official assertions or diagnostics are passed to the model. Inspect only operational metadata until all supplemental outputs are frozen. The other 55 tasks remain unopened.", + "runtime_environment": "UV_PROJECT_ENVIRONMENT=/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/dsh-harbor-venv", + "status_at_registration": "Initial direct-download preflight stopped at the completed second-pair boundary: three passed, one setup timeout, two not started. No model executions. This v2 registration requires all six fresh preflights with verified wheel transport.", + "historical_pi_results": "benchmarks/harbor/results/pi-deepseek-flash-dev34-20261024.json", + "authorization": { + "kind": "separately authorized fresh supplemental attempts", + "requested_at_date": "2026-11-01", + "scope": "Rerun all six analytically unscored tasks from the first official Harness dev34 run. Keep both attempts. Prior authorization to use the configured official DeepSeek credential remains applicable.", + "original_results": "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json", + "original_results_sha256": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396", + "original_freeze": "benchmarks/harbor/configs/deepseek-harness-dev34-20261030/outputs-freeze.json", + "original_freeze_sha256": "cc421c1429450ac41d355d0df8333f4ab10ca4da0bf47f24bc3892d0724d9c24" + }, + "transport_amendment": "Use UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple to forward the current original PyPI index, preserving versions, requirements, yanks, hashes and extra-index settings. Serve original wheels only after checking current PyPI release filename/size/SHA256/HTTPS origin and full downloaded bytes. Existing ranged downloader, cross-process file locks, and persistent public provenance log. Set UV_HTTP_TIMEOUT=600 for a cold wheel to finish verification before any body is served. Setup remains 1800 seconds; task-native agent/verifier limits and official scripts are unchanged. Original APT and uv/Python release transports remain identical. No model output or task is replayed automatically.", + "supersedes_model_queue": { + "plan": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json", + "sha256": "cb45a79c8c0b128d6807118a656afc32357de40c250bb6064b9d46798a4232ac", + "model_jobs_started": 0, + "initial_preflight_evidence": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json", + "initial_preflight_evidence_sha256": "ead4b47154c9507326f6f03eeada33a2eb486ab027317fc3b8d655df7bd4a7c9", + "policy": "Preserve all initial preparation evidence. Neither initial model queue nor initial third preflight pair will run. All six v2 preflights must pass before the six new formal attempts." + }, + "transport_audit_log": ".cache/nanopy-harbor/pypi-ranges/audit/dsh-sup6-cache-20261101.jsonl", + "transport_source_ref": "dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3" +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json new file mode 100644 index 0000000..151206b --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json @@ -0,0 +1,136 @@ +{ + "schema_version": 1, + "audited_at": "2026-11-01T15:26:59.206575+00:00", + "original_plan": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json", + "sha256": "cb45a79c8c0b128d6807118a656afc32357de40c250bb6064b9d46798a4232ac" + }, + "summary": { + "installation_passed": 3, + "installation_failed": 1, + "installation_not_started": 2, + "model_executions": 0, + "official_verifier_executions": 0 + }, + "reason": "PyTorch model recovery reached the 1800-second setup limit while downloading original verifier dependencies. After pair 2 completed, stop the controller before pair 3 and register verified PyPI wheel transport. No running trial was interrupted by this boundary stop.", + "boundary_stop": { + "path": "jobs/dsh-supplement-preflight-boundary-stop.json", + "sha256": "1a058982408952541e9a98968d4c7e85f36ba050fe6693b1ef994b08ab3b5800" + }, + "controller_exit_code": 130, + "active_task_containers_after_stop": 0, + "trials": [ + { + "task": "terminal-bench/multi-source-data-merger", + "trial": "multi-source-data-merger__A9ZBWfv", + "status": "setup_passed", + "exception_type": null, + "verifier_profile": { + "profile": "multi-source-data-merger-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T14:37:39.038730Z", + "finished_at": "2026-11-01T14:39:16.479499Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-01-20261101.json", + "sha256": "7907d0ca8916a3f8ac35a763309bf13eef2e06e3a9477f4353d8ef1d2450f79e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/result.json", + "sha256": "c3efdc8bb4fb1048f32aa4c88105b8c1a08262e98c2fbe329d41dd09853ae408" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/agent/bootstrap.json", + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-cli", + "trial": "pytorch-model-cli__WHD5rDb", + "status": "setup_passed", + "exception_type": null, + "verifier_profile": { + "profile": "pytorch-model-cli-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T14:37:39.001514Z", + "finished_at": "2026-11-01T14:52:46.672756Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-01-20261101.json", + "sha256": "7907d0ca8916a3f8ac35a763309bf13eef2e06e3a9477f4353d8ef1d2450f79e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/result.json", + "sha256": "1061d55f449577a08bfb866a4572bf1cda8efeba9e58e731b5d70a7972df0d22" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/agent/bootstrap.json", + "sha256": "a3c0057981c5d3db5eefa64571d2bcc3354657ae9602961506ab91df8197e94a" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "trial": "pytorch-model-recovery__LD8MzLU", + "status": "setup_failed", + "exception_type": "AgentSetupTimeoutError", + "verifier_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "incomplete" + }, + "agent_setup": { + "started_at": "2026-11-01T14:53:24.216340Z", + "finished_at": "2026-11-01T15:23:24.570179Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-02-20261101.json", + "sha256": "61fa3336d6839abc032ab09e74b840c8b17301726df6d2955dfef9bb58eccd82" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/result.json", + "sha256": "57b535c90e77e1bfb129bdb32e45f29e825132e0dfde847fcab2be97346db9b3" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/agent/bootstrap.json", + "sha256": "569634e2774a66f6a021d30250a11667067e2fb7503586e9cffd6f8eadacdb75" + } + ] + }, + { + "task": "terminal-bench/query-optimize", + "trial": "query-optimize__nZVwCmT", + "status": "setup_passed", + "exception_type": null, + "verifier_profile": { + "profile": "query-optimize-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T14:53:24.209263Z", + "finished_at": "2026-11-01T14:54:26.697416Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-02-20261101.json", + "sha256": "61fa3336d6839abc032ab09e74b840c8b17301726df6d2955dfef9bb58eccd82" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/result.json", + "sha256": "bcddf59e8ca8e1042bc40c41d2928ef8a7ec7d419cc2fe234f42ae00ba304de2" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/agent/bootstrap.json", + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826" + } + ] + } + ] +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/source-comparison.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/source-comparison.json new file mode 100644 index 0000000..f3fc7be --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/source-comparison.json @@ -0,0 +1,17 @@ +{ + "checked_at": "2026-11-01T14:47:03.295770+00:00", + "original_adapter_ref": "5b503ab6d04af2f6a719ec0dbe5418bfad4bf128", + "supplement_adapter_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "changed_adapter_files": [ + "benchmarks/harbor/src/harbor_adapter/bootstrap.py" + ], + "identical_runtime_files_sha256": { + "deepseek.py": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "_dsh_client.py": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "_agent_runner.py": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "verifier.py": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "_verifier_runner.py": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "adapter.py": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02" + }, + "interpretation": "Only bootstrap dependency-profile definitions changed in adapter source. Harness driver, prompt delivery, model settings, agent process supervision and verifier implementation are byte-identical. The separately registered task environment changes UV_HTTP_TIMEOUT from the default 30 to 120 seconds." +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json new file mode 100644 index 0000000..b88002a --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json @@ -0,0 +1,148 @@ +{ + "services": { + "main": { + "volumes": [ + { + "type": "bind", + "source": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf", + "target": "/etc/apt/apt.conf.d/99nanopy-index-cache", + "read_only": true + } + ], + "environment": { + "UV_INSTALLER_GITHUB_BASE_URL": "http://127.0.0.1:18743", + "UV_PYTHON_INSTALL_MIRROR": "http://127.0.0.1:18743/astral-sh/python-build-standalone/releases/download", + "UV_HTTP_TIMEOUT": "600", + "UV_DEFAULT_INDEX": "http://127.0.0.1:18744/simple" + } + }, + "apt-index-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/apt_index_proxy.py", + "--bind", + "127.0.0.1", + "--port", + "18742", + "--cache", + "/cache", + "--workers", + "16" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "mem_limit": "256m", + "cpus": 0.5, + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/apt_index_proxy.py:/proxy/apt_index_proxy.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/apt-indexes:/cache" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18742), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + }, + "uv-release-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/uv_release_cache.py", + "serve", + "--cache", + "/cache", + "--manifest", + "/manifest.json" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "cpus": 0.5, + "mem_limit": "256m", + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/uv_release_cache.py:/proxy/uv_release_cache.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/uv-releases:/cache:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json:/manifest.json:ro" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18743), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + }, + "pypi-wheel-cache": { + "image": "alexgshaw/pytorch-model-cli@sha256:c996884b5fb5fb2c0c07d67da2c4f974f271f07234ed28e1a868240375c57a67", + "network_mode": "service:main", + "entrypoint": [ + "python3", + "/proxy/pypi_wheel_cache.py", + "--cache", + "/cache", + "--bind", + "127.0.0.1", + "--port", + "18744", + "--workers", + "16", + "--downloads", + "2", + "--audit-log", + "/cache/audit/dsh-sup6-cache-20261101.jsonl" + ], + "user": "1000:1000", + "read_only": true, + "cap_drop": [ + "ALL" + ], + "security_opt": [ + "no-new-privileges:true" + ], + "mem_limit": "512m", + "cpus": 1.0, + "volumes": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/pypi_wheel_cache.py:/proxy/pypi_wheel_cache.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/scripts/cache_image.py:/proxy/cache_image.py:ro", + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/.cache/nanopy-harbor/pypi-ranges:/cache" + ], + "healthcheck": { + "test": [ + "CMD", + "python3", + "-c", + "import socket; socket.create_connection((\"127.0.0.1\",18744), timeout=2).close()" + ], + "interval": "1s", + "timeout": "3s", + "retries": 30 + } + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json new file mode 100644 index 0000000..b1d3e8e --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + }, + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json new file mode 100644 index 0000000..ffe4149 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + }, + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json new file mode 100644 index 0000000..71e5853 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json @@ -0,0 +1,42 @@ +{ + "n_concurrent_trials": 2, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + }, + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json new file mode 100644 index 0000000..f0cb964 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/multi-source-data-merger", + "ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json new file mode 100644 index 0000000..ad46d3b --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-cli", + "ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json new file mode 100644 index 0000000..ca8b322 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/pytorch-model-recovery", + "ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json new file mode 100644 index 0000000..7971f70 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/query-optimize", + "ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json new file mode 100644 index 0000000..df45c36 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-pipeline-parallelism", + "ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json new file mode 100644 index 0000000..ac6bcf6 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json @@ -0,0 +1,38 @@ +{ + "n_concurrent_trials": 1, + "n_attempts": 1, + "retry": { + "max_retries": 0 + }, + "agents": [ + { + "name": "harbor_adapter:DeepSeekHarnessAgent", + "model_name": "deepseek/deepseek-flash", + "override_setup_timeout_sec": 1800, + "kwargs": { + "distribution_path": "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/deepseek-harness-build-f3803c9/distribution", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "reasoning_effort": "high" + } + } + ], + "environment": { + "type": "docker", + "extra_docker_compose": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json" + ] + }, + "tasks": [ + { + "name": "terminal-bench/torch-tensor-parallelism", + "ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684" + } + ], + "verifier": { + "import_path": "harbor_adapter:RetryingVerifier", + "kwargs": { + "max_attempts": 3, + "require_agent_quiescence": true + } + } +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json new file mode 100644 index 0000000..ec58019 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json @@ -0,0 +1,71 @@ +{ + "recorded_at": "2026-11-01T15:28:42.170928+00:00", + "range_download": { + "filename": "torch-2.7.1-cp313-cp313-manylinux_2_28_x86_64.whl", + "path": ".cache/nanopy-harbor/pypi-ranges/blobs/sha256/d632f5417b6980f61404a125b999ca6ebd0b8b4bbdbb5fbbba44374ab619a412", + "size": 821033192, + "sha256": "d632f5417b6980f61404a125b999ca6ebd0b8b4bbdbb5fbbba44374ab619a412", + "elapsed_seconds": 357.6746226220057, + "bytes_per_second": 2295475.105226228, + "source": "https://pypi.org/pypi/torch/2.7.1/json", + "purpose": "Host-side verified transport probe; not injected into the active preflight." + }, + "initial_native_uv_smoke": { + "checked_at": "2026-11-01T15:19:10.795930+00:00", + "command": [ + "jobs/uv-0.9.5-probe/uv", + "tool", + "run", + "--python", + "/home/minix/.local/share/uv/python/cpython-3.13.15-linux-x86_64-gnu/bin/python3.13", + "--with", + "pytest==8.4.1", + "--with", + "pytest-json-ctrf==0.3.5", + "pytest", + "--version" + ], + "uv_version": "0.9.5", + "script_sha256": "8369dfbba73fb6410b9dbef79d7ddcd4d1190fef90fbf35d58a91ea41f53fbb1", + "exit_code": 0, + "stdout": "pytest 8.4.1", + "model_calls": 0, + "fresh_uv_cache": true, + "default_index": "Loopback proxy forwarding current original PyPI metadata and verified wheels." + }, + "final_native_uv_smoke": { + "checked_at": "2026-11-01T15:26:24.601919+00:00", + "command": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/uv-0.9.5-probe/uv", + "tool", + "run", + "--python", + "/home/minix/.local/share/uv/python/cpython-3.13.15-linux-x86_64-gnu/bin/python3.13", + "--with", + "pytest==8.4.1", + "--with", + "pytest-json-ctrf==0.3.5", + "pytest", + "--version" + ], + "uv_version": "0.9.5", + "script_sha256": "ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74", + "probe_sha256": "02d43b929e97cb8a73c32525203cb9fa439294aa3dbfb55e804a41c009483237", + "exit_code": 0, + "stdout": "pytest 8.4.1", + "model_calls": 0, + "fresh_uv_cache": true, + "audit_log": "jobs/dsh-pypi-uv-smoke-audit-events.jsonl", + "audit_log_sha256": "330958a0b95493071fc429c43ed8c4f25a96ff648b88ff4531dc29fd9976b796", + "default_index": "Loopback proxy forwarding current original PyPI metadata and verified wheels." + }, + "validation": { + "full_harbor_tests_passed": 301, + "pypi_cache_tests_passed": 16 + }, + "source_ref": "dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3", + "source_files_sha256": { + "benchmarks/harbor/scripts/pypi_wheel_cache.py": "ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74", + "benchmarks/harbor/scripts/cache_image.py": "11ddde3fe86a6e4bfdefe7c3578794df98d237f701e40f219d76a541911c9f0c" + } +} From 7f1fed6a96b90ef6db3406f9c56f4a115c00028e Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 07:50:02 -0800 Subject: [PATCH 13/15] test(benchmarks): record six passed supplemental preflights --- .../README.md | 7 +- .../launch-evidence.json | 22 ++ .../preflight-evidence.json | 255 ++++++++++++++++++ 3 files changed, 282 insertions(+), 2 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md index 803416b..628485e 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md @@ -1,8 +1,11 @@ # Authorized six-task DeepSeek Harness supplement Status: the [v2 registration](experiment-v2.json) adds verified PyPI wheel -transport after the initial preparation encountered a setup timeout. Fresh v2 -dependency preflight is pending; no supplemental model executions have started. +transport after the initial preparation encountered a setup timeout. All six fresh v2 +dependency checks passed. [Preflight evidence](preflight-evidence.json) and +[launch registration](launch-evidence.json) precede the six formal model jobs. +Formal execution is now registered; results will be audited after all outputs +are frozen. The initial direct-download preflight passed data merger, PyTorch CLI, and query optimization. Model recovery exceeded its 1,800-second setup limit while diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json new file mode 100644 index 0000000..ad48bfb --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json @@ -0,0 +1,22 @@ +{ + "schema_version": 1, + "authorized_launch_registered_at": "2026-11-01T15:49:52.188939+00:00", + "repository_ref": "0c587c016fc0757f08b223a1976fae9e800d352b", + "adapter_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "transport_ref": "dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3", + "authorization": "The maintainer requested rerunning the six unscored tasks on 2026-11-01. Prior explicit confirmation authorizes use of the existing DeepSeek provider credential without publishing it.", + "phase": "model", + "model_jobs_started_at_registration": 0, + "model_runs_authorized": 6, + "whole_trial_retries": 0, + "preflight": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json", + "sha256": "e973e6d280171287c8c60b7888834b56259f897aac67e8ffa6c23daad39e3e06" + }, + "plan": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json", + "sha256": "69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93" + }, + "credential_handling": "Read the authorized local DeepSeek provider credential into the child environment only. No key is written to this record.", + "output_policy": "Freeze all six outputs before inspecting new model trajectories or hidden assertions." +} diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json new file mode 100644 index 0000000..95640dc --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json @@ -0,0 +1,255 @@ +{ + "schema_version": 1, + "audited_at": "2026-11-01T15:49:52.094028+00:00", + "experiment": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json", + "sha256": "69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93" + }, + "adapter_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "model_jobs_started": 0, + "summary": { + "installation_passed": 6, + "sdk_initialization_passed": 6, + "supervision_probe_passed": 6, + "verifier_dependency_profiles_passed": 6, + "verifier_dependency_profiles_not_configured": 0, + "model_executions": 0 + }, + "trials": [ + { + "task": "terminal-bench/multi-source-data-merger", + "task_ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "job": "dsh-sup6-install-pair-01-cache-20261101", + "trial": "multi-source-data-merger__kDiQNmh", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "multi-source-data-merger-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:30:37.424635Z", + "finished_at": "2026-11-01T15:32:11.448362Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-01-cache-20261101.json", + "sha256": "866efbc2def799a1e374e0e47e4b6f596f2b9c7596525b0c53989afab2656fa0" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/manifest.json", + "sha256": "a1c0a37704126265f30150d7ab8d049adcbef97964b3ca57cc64c927dec3c50e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/summary.json", + "sha256": "7e60fc4d3c5a8ac06c48f630b8359cce05366334fd479d196c49e518f824ed56" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/result.json", + "sha256": "c2457b2afd8de269bfac9c2c67a0c120d1f2604bee58da87f3f01fef3bac4e37" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/agent/bootstrap.json", + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-cli", + "task_ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "job": "dsh-sup6-install-pair-01-cache-20261101", + "trial": "pytorch-model-cli__dzNhbDJ", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "pytorch-model-cli-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:30:37.388855Z", + "finished_at": "2026-11-01T15:34:17.660530Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-01-cache-20261101.json", + "sha256": "866efbc2def799a1e374e0e47e4b6f596f2b9c7596525b0c53989afab2656fa0" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/manifest.json", + "sha256": "a1c0a37704126265f30150d7ab8d049adcbef97964b3ca57cc64c927dec3c50e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/summary.json", + "sha256": "7e60fc4d3c5a8ac06c48f630b8359cce05366334fd479d196c49e518f824ed56" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/result.json", + "sha256": "942b30020eaf1fb96c31f6b9ff272eaf32364f08666f2ff3cb4388e4d089d486" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/agent/bootstrap.json", + "sha256": "0d3b19888676a0f8c1bd7b850d8a349a49b07f1cebc2e1780e678948be935a60" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "task_ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "job": "dsh-sup6-install-pair-02-cache-20261101", + "trial": "pytorch-model-recovery__XnktXzC", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:34:54.988540Z", + "finished_at": "2026-11-01T15:38:52.082773Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-02-cache-20261101.json", + "sha256": "85ad71a6a8dd6e53fd18764b3eb819e725a00d1fb6f640e130d583aa095a3678" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/manifest.json", + "sha256": "bf7a0c43b592960af76176b8715b8440129886b6becc68f51dce7c2dafb06bcf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/summary.json", + "sha256": "40582b6a4c73c6f9ad84f1e454e3072b5818e5b4213e08082f2b88a3b88f0aa1" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/result.json", + "sha256": "ef17615f7c6a6c91d075791d90ad56adab7e245eb7bda5ce846bbfbf904f2366" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/agent/bootstrap.json", + "sha256": "638922b88c59516844637223aff41c51866cf5e21824c22d7fd7116ebfb1e0fc" + } + ] + }, + { + "task": "terminal-bench/query-optimize", + "task_ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "job": "dsh-sup6-install-pair-02-cache-20261101", + "trial": "query-optimize__mjkeDjx", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "query-optimize-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:34:54.987141Z", + "finished_at": "2026-11-01T15:35:49.586138Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-02-cache-20261101.json", + "sha256": "85ad71a6a8dd6e53fd18764b3eb819e725a00d1fb6f640e130d583aa095a3678" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/manifest.json", + "sha256": "bf7a0c43b592960af76176b8715b8440129886b6becc68f51dce7c2dafb06bcf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/summary.json", + "sha256": "40582b6a4c73c6f9ad84f1e454e3072b5818e5b4213e08082f2b88a3b88f0aa1" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/result.json", + "sha256": "f1e90e69169bf7cfee37a68f7915b981e4cff370caa457d9ba9dffaaa431b7cf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/agent/bootstrap.json", + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826" + } + ] + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "task_ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "job": "dsh-sup6-install-pair-03-cache-20261101", + "trial": "torch-pipeline-parallelism__mXeoYMQ", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "torch-pipeline-parallelism-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:39:41.683509Z", + "finished_at": "2026-11-01T15:48:44.616122Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-03-cache-20261101.json", + "sha256": "2ea8ddedb98188185000466de388c6d4c62c565ebb5a6d457b074ac1adad86b3" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/manifest.json", + "sha256": "a79d0cfc3f8c90c299dfdbf2f245115c51924cb56026116b53ccd54bc44abf9d" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/summary.json", + "sha256": "6ac52365e43f74afd60a3424faf8d05b7df928abeca72c1559362664ecd8b348" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/result.json", + "sha256": "d29a589b7595b286187ff3dea2ec985151dfda8cb19da12fee759f75c59771bf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/agent/bootstrap.json", + "sha256": "a9460a42b2e140bbe0bee9d4a2226d95a34d556d4cdf92a6b310b8b3a72c3bca" + } + ] + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "task_ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "job": "dsh-sup6-install-pair-03-cache-20261101", + "trial": "torch-tensor-parallelism__6VSbzJS", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "torch-tensor-parallelism-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:39:41.684726Z", + "finished_at": "2026-11-01T15:48:43.633194Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-03-cache-20261101.json", + "sha256": "2ea8ddedb98188185000466de388c6d4c62c565ebb5a6d457b074ac1adad86b3" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/manifest.json", + "sha256": "a79d0cfc3f8c90c299dfdbf2f245115c51924cb56026116b53ccd54bc44abf9d" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/summary.json", + "sha256": "6ac52365e43f74afd60a3424faf8d05b7df928abeca72c1559362664ecd8b348" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/result.json", + "sha256": "4862f8fa3a7b1038b169aa97a54f0892917ad45bdb769a3fb25fba62ade08979" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/agent/bootstrap.json", + "sha256": "34af5d820280f7c297e211a7e512cc9101f09ab3908a0b7bf5ded9486eb61085" + } + ] + } + ] +} From 90191f66b187b4f17ecdc2f99ee0bc59dd2e0309 Mon Sep 17 00:00:00 2001 From: minixalpha Date: Sun, 1 Nov 2026 09:03:42 -0800 Subject: [PATCH 14/15] test(benchmarks): record scored deepseek supplement results --- benchmarks/harbor/README.md | 7 + benchmarks/harbor/README.zh-CN.md | 6 + .../README.md | 19 +- .../outputs-freeze.json | 2372 +++++++++++++++++ .../en/deepseek-harness-dev34-20261030.md | 4 + ...epseek-harness-dev6-supplement-20261101.md | 161 ++ .../zh-CN/deepseek-harness-dev34-20261030.md | 3 + ...epseek-harness-dev6-supplement-20261101.md | 142 + ...seek-harness-dev6-supplement-20261101.json | 2054 ++++++++++++++ 9 files changed, 4763 insertions(+), 5 deletions(-) create mode 100644 benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json create mode 100644 benchmarks/harbor/reports/en/deepseek-harness-dev6-supplement-20261101.md create mode 100644 benchmarks/harbor/reports/zh-CN/deepseek-harness-dev6-supplement-20261101.md create mode 100644 benchmarks/harbor/results/deepseek-harness-dev6-supplement-20261101.json diff --git a/benchmarks/harbor/README.md b/benchmarks/harbor/README.md index 5191fd1..932834b 100644 --- a/benchmarks/harbor/README.md +++ b/benchmarks/harbor/README.md @@ -228,6 +228,13 @@ The report retains those failures and documents the [keyless lifecycle probe](scripts/probe_deepseek_lifecycle.py). Treat these results as measurements of the registered SDK configuration. +The separately authorized [six-task supplement](reports/en/deepseek-harness-dev6-supplement-20261101.md) +reran the unscored tasks with reviewed dependency preflight and verified original +PyPI wheel transport: four passed, two failed, and none remained unscored. The +original results stay unchanged. Combining the 28 original accepted attempts +with these six new attempts gives mixed coverage of 26 passes and eight failures; +this outcome-selected supplement is not an independent benchmark. + ### Cache original task images If Docker image pulls repeatedly stall, the optional cache utility downloads the diff --git a/benchmarks/harbor/README.zh-CN.md b/benchmarks/harbor/README.zh-CN.md index 3ca5d75..0d69135 100644 --- a/benchmarks/harbor/README.zh-CN.md +++ b/benchmarks/harbor/README.zh-CN.md @@ -182,6 +182,12 @@ Python 3.13 和冻结的 wheel 集合。仅安装检查会初始化真实 SDK [无密钥生命周期探针](scripts/probe_deepseek_lifecycle.py)。这些结果衡量的是 注册的 SDK 配置。 +另行授权的[六题补测](reports/zh-CN/deepseek-harness-dev6-supplement-20261101.md) +使用经过审核的依赖预检和经过校验的原始 PyPI wheel 下载通道,重跑未计分题目: +四题通过、两题失败,无剩余未计分项。原始结果保持不变。将原始 28 次有效尝试 +与这六次新尝试结合,得到 26 题通过、八题失败的混合覆盖;这种根据原始结果 +选取的补测不构成独立基准。 + ### 缓存原始任务镜像 如果 Docker 拉取镜像反复停滞,可用缓存工具从同一 Docker Hub 公共仓库分段并行下载: diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md index 628485e..2ef6c48 100644 --- a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/README.md @@ -1,11 +1,20 @@ # Authorized six-task DeepSeek Harness supplement -Status: the [v2 registration](experiment-v2.json) adds verified PyPI wheel -transport after the initial preparation encountered a setup timeout. All six fresh v2 -dependency checks passed. [Preflight evidence](preflight-evidence.json) and +Status: completed and audited on November 1, 2026: **four passes, two scored +failures, and zero unscored errors**. All six model executions completed normally; +all official verifiers finished on the first attempt. CLI failed prediction +comparisons, and pipeline parallelism failed two backward-value comparisons. +See the [full report](../../reports/en/deepseek-harness-dev6-supplement-20261101.md) +and [structured results](../../results/deepseek-harness-dev6-supplement-20261101.json). + +The [v2 registration](experiment-v2.json) adds verified PyPI wheel transport after +the initial preparation encountered a setup timeout. All six fresh dependency +checks passed. [Preflight evidence](preflight-evidence.json) and [launch registration](launch-evidence.json) precede the six formal model jobs. -Formal execution is now registered; results will be audited after all outputs -are frozen. +The [freeze manifest](outputs-freeze.json) covers 590 files, frozen before reviewing +new trajectories or hidden assertions. All 2,506 original frozen files remain +unchanged. The 28 original accepted attempts plus these six new attempts give +mixed coverage of 26 passes and eight failures, not an independent clean run. The initial direct-download preflight passed data merger, PyTorch CLI, and query optimization. Model recovery exceeded its 1,800-second setup limit while diff --git a/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json new file mode 100644 index 0000000..b2051d4 --- /dev/null +++ b/benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json @@ -0,0 +1,2372 @@ +{ + "schema_version": 1, + "frozen_at": "2026-11-01T16:56:07.773103+00:00", + "registered_model_jobs": 6, + "installation_trials": 10, + "initial_preflight_trials": 4, + "final_preflight_trials": 6, + "policy": "All registered jobs finished before this freeze. No new solution trajectories or hidden test contents were reviewed before the freeze. Raw logs remain local; this manifest publishes paths, sizes, and hashes only.", + "file_count": 590, + "files": { + ".cache/nanopy-harbor/pypi-ranges/audit/dsh-sup6-cache-20261101.jsonl": { + "sha256": "ef8385be4d1387b72014dca1e89b1281fed347714f9bd3e5f9e050b501e7366a", + "size": 405336 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/apt-cache.conf": { + "sha256": "d0e1ffb6dbd02c54978ba2df386326311a59f6d3774990180f22bcc6adfaedbc", + "size": 549 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/deepseek-distribution.json": { + "sha256": "34410bdce15ec3b417105eb7f9f6e01a14c47f1bfd0ffe16d76338e7efa51dd2", + "size": 2742 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-cache.compose.json": { + "sha256": "560bc83f2a88bb907662db1b60ab15f49443a1572822fa633440f3cc76510557", + "size": 3317 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/dependency-release-manifest.json": { + "sha256": "77dfda40e45b870a8ce7f07a1e058d416d94a04ab3d27c2f21bc370f22151394", + "size": 3250 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json": { + "sha256": "69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93", + "size": 24581 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json": { + "sha256": "cb45a79c8c0b128d6807118a656afc32357de40c250bb6064b9d46798a4232ac", + "size": 20403 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/image-digests.json": { + "sha256": "c2405e16c2aaf2d837b6f6594875e197fd13eb94ed094253e4513c315783d030", + "size": 2488 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-01.json": { + "sha256": "cc4f07cff8a7ec9ffe4de4bdb69322b0ec89b62c72072020d1f9b1f7f0e20e26", + "size": 1268 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-02.json": { + "sha256": "8c918a0998feb41ac3a974228d1bc3b4ba220fb5d76283c77590755738c1d149", + "size": 1263 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/install-pair-03.json": { + "sha256": "b04b132fa453bec7c4f888e6e52dab533b8870dc1c0ea21881fb887c84ac00ba", + "size": 1277 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json": { + "sha256": "efdf89be02eda588ef14952ae1aa9844c483f1060136ea5b05bb9b8f5833c943", + "size": 1282 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json": { + "sha256": "452c5114c9ac5fc82634ce25ce8a2ee0a8f4e3ee15724f5324bdb1befcf865c3", + "size": 489 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/multi-source-data-merger.json": { + "sha256": "4e25ecef8ce9beb5a0adfb9977242fe32a9a5559b8f691fa686971d4a6d4e4e1", + "size": 1118 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json": { + "sha256": "e973e6d280171287c8c60b7888834b56259f897aac67e8ffa6c23daad39e3e06", + "size": 12105 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json": { + "sha256": "ead4b47154c9507326f6f03eeada33a2eb486ab027317fc3b8d655df7bd4a7c9", + "size": 5773 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-cli.json": { + "sha256": "a0805861bc826d29e2bc9205c2c720f725f52a549d970c61b407f786dd73df49", + "size": 1111 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/pytorch-model-recovery.json": { + "sha256": "fddffa4e9c5bac957f87460cdd04276f964fb4c2aa1bcd98124387d95f140c99", + "size": 1116 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/query-optimize.json": { + "sha256": "2320f05a860f4c91d1fd1a200740cda5ac8fa67ea65dfab3dc3a5b0ef745b6e0", + "size": 1108 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/selection.json": { + "sha256": "5047e6d007ed45ee6887bcf09cceb3475b7075e56551a43143c38b8e00991284", + "size": 695 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/source-comparison.json": { + "sha256": "2a9136e82334bea1c4d7a1097537d36227d3e63836a4fc9983709d288e4f749c", + "size": 1188 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-pipeline-parallelism.json": { + "sha256": "bfa979780e67235b1f5472d025ca7b8cb2ea2b4ef2c7f1fab29d5b4fabee8d5c", + "size": 1120 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/torch-tensor-parallelism.json": { + "sha256": "11b91bacfe4d1b2fdcf57a23376ae61b16df9c69802bd5465557db95b95976d8", + "size": 1118 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/dependency-cache.compose.json": { + "sha256": "b057be706c6139af66a7a460e8853ba83138986a2424b957753dc832e2f265a8", + "size": 4781 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-01.json": { + "sha256": "a2cdbfd25baa068e6f904fc6843a4657991d5a6c8fb8a7b16ee18ffeba58e9dd", + "size": 1281 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-02.json": { + "sha256": "8260804d67294d30901161efed81d18d6263ab5669e124ae3ea495efd05dbee0", + "size": 1276 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/install-pair-03.json": { + "sha256": "a2c56775c102aa0acce248d3ca5bf65c70e2c94b3b2805fe738b37e3c461d798", + "size": 1290 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/multi-source-data-merger.json": { + "sha256": "fb57f5183e4f4922815881ba050583175238cc14cf5eb766e7ade1e53f1aaf24", + "size": 1131 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-cli.json": { + "sha256": "ab45588a2c978419e38ebd5aec8cd2b80a8946393e2931930a145554c743b226", + "size": 1124 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/pytorch-model-recovery.json": { + "sha256": "efb3470630dd176e013bdf739354e23fb640bdd086b9ce17e67d79d8fa71efa5", + "size": 1129 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/query-optimize.json": { + "sha256": "253c2b9bd43cdc526cddb1116f4d3a2b27bcd277c73ad0a794fe744120fd841b", + "size": 1121 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-pipeline-parallelism.json": { + "sha256": "d1228591c9e282dbf36effdb3589efed0ea711007be9d7a09d35025a35f54131", + "size": 1133 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/torch-tensor-parallelism.json": { + "sha256": "df3772684d82ca3a55a682f855f28679b8f4d12d46e182bdfdb0e2cb50f66a61", + "size": 1131 + }, + "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json": { + "sha256": "d734625b2ba1e5624a31ab5164c3bbe8c1b78c60377a366ebe4a9cf4f239569f", + "size": 2769 + }, + "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json": { + "sha256": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396", + "size": 263402 + }, + "benchmarks/harbor/scripts/apt_index_proxy.py": { + "sha256": "bc15f7adf1d7968d627ceca4683ab86742a7188a1b1955ba1fef37ba48e8a936", + "size": 10327 + }, + "benchmarks/harbor/scripts/cache_image.py": { + "sha256": "11ddde3fe86a6e4bfdefe7c3578794df98d237f701e40f219d76a541911c9f0c", + "size": 17103 + }, + "benchmarks/harbor/scripts/pypi_wheel_cache.py": { + "sha256": "ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74", + "size": 10623 + }, + "benchmarks/harbor/scripts/run_comparison.py": { + "sha256": "228dbff33c0ec8a2359e612da632ef543dcac4273db2f9a7d0573a24e856fbae", + "size": 18201 + }, + "benchmarks/harbor/scripts/uv_release_cache.py": { + "sha256": "993cc35b1f3a9e8e171b662ab6d4524c5fe059f0d1f0e275a5bb1930111e0ff5", + "size": 10774 + }, + "jobs/deepseek-alias-docs-20261101.html": { + "sha256": "723bbdcf0b8fd45ae629b0c9fa3939bbfa2af368c69c2ce37ff532049ddde9e3", + "size": 51222 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-01-20261101.json": { + "sha256": "7907d0ca8916a3f8ac35a763309bf13eef2e06e3a9477f4353d8ef1d2450f79e", + "size": 135197 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-01-20261101.log": { + "sha256": "962506f1763db5c0dabcf13b4170595ac7dacf3613e52bb9f7060f25fe35cee8", + "size": 915 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-02-20261101.json": { + "sha256": "61fa3336d6839abc032ab09e74b840c8b17301726df6d2955dfef9bb58eccd82", + "size": 45082 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-02-20261101.log": { + "sha256": "5f9e63160b50a1acfea2c177f99aa1f5c7802548f2d52eccf0895b0e05687a41", + "size": 293 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/image-cache.jsonl": { + "sha256": "00aca15594721e0a548727b164a89affb9166735a12e8cc5fc96a3001db7a2b0", + "size": 1011 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/image-downloads.log": { + "sha256": "ce40e3210825e83307678ede1413f290fd1c11f87d8fe9d5dd60ed0df17dfdbe", + "size": 1665 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/initial-docker-images.json": { + "sha256": "449d13f34e0b28d37629f60676696304188487bd8f7361f38ffae2ebbe0b4f60", + "size": 465 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/registration.json": { + "sha256": "306e0511feffaa093584ed765cfcf90c779b35f6a3c4c7eb684070e33f2d7079", + "size": 21140 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-01-cache-20261101.json": { + "sha256": "866efbc2def799a1e374e0e47e4b6f596f2b9c7596525b0c53989afab2656fa0", + "size": 135241 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-01-cache-20261101.log": { + "sha256": "22703e1016c018d6c9f3c39d944c94a915569bced2d50e59490571035b1cc281", + "size": 366 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-02-cache-20261101.json": { + "sha256": "85ad71a6a8dd6e53fd18764b3eb819e725a00d1fb6f640e130d583aa095a3678", + "size": 45046 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-02-cache-20261101.log": { + "sha256": "94e999a78facb25a79ff16c1794fa735fc610e6c33de46151de3fdd460eb3c2c", + "size": 305 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-03-cache-20261101.json": { + "sha256": "2ea8ddedb98188185000466de388c6d4c62c565ebb5a6d457b074ac1adad86b3", + "size": 56211 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-03-cache-20261101.log": { + "sha256": "be2dd6ca243317fdb656dea5a9f08191fd689912f5fcd1ef465d229cd7420c69", + "size": 305 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-multi-source-data-merger-cache-20261101.json": { + "sha256": "19beb040db143599f69d5bec8e7d8603d904967a39dfd61af052a774ca3b82e1", + "size": 25153 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-multi-source-data-merger-cache-20261101.log": { + "sha256": "b88e146eb020d7a830d39ae25dc0e4aeff177788316e083deba8842785cf6b0a", + "size": 384 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-cli-cache-20261101.json": { + "sha256": "60351fe47c843efb3c964a39b2c46e1fe247ef0feaa47f98a7a17c1b1d38d354", + "size": 112275 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-cli-cache-20261101.log": { + "sha256": "921baf4f1871e9c1ad13a313f17e293752904499cf99af883277be692dd63e88", + "size": 309 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-recovery-cache-20261101.json": { + "sha256": "be0a7e21e0388145ea0d1dc8efcb9f44a30c91c4f98f5f6edee80893626afbc1", + "size": 27480 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-recovery-cache-20261101.log": { + "sha256": "e6fcc64fee3ec472a9b1e1454c678742600da1daac2ca65901c0c2a1fde77085", + "size": 319 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-query-optimize-cache-20261101.json": { + "sha256": "fb1e727669d4c01a13753ae5122f3c8b1d433c221b32e0c317bb31d1e7fac9fa", + "size": 19725 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-query-optimize-cache-20261101.log": { + "sha256": "3b37095f25d27b2f5f30e631c5b70cb10671ac4dfb012e6c8cf6d38d0600de60", + "size": 303 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-pipeline-parallelism-cache-20261101.json": { + "sha256": "e2ff45301e23a775b6e044d69be6c4e62c3f9304769381512d5fce3b80ff9cce", + "size": 29236 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-pipeline-parallelism-cache-20261101.log": { + "sha256": "4d7a8da6dadeeb419eddafc02dec2992906b5e1066d292420f955afa7f86dd4b", + "size": 327 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-tensor-parallelism-cache-20261101.json": { + "sha256": "8f0d3027d7c1be04eb4453025fa7638dc25f8be7d16f8db8bc55573868673333", + "size": 29218 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-tensor-parallelism-cache-20261101.log": { + "sha256": "02a3580bbb8f107bef679b2c783eddf0d377b88002a8fd782e566c4fd2783933", + "size": 323 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/image-cache.jsonl": { + "sha256": "36a76b09df6b05e5a680bd661c2cf8dd63ff4d90c227c28b0d9d29cc5b6351fc", + "size": 4920 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/image-downloads.log": { + "sha256": "7082c78fb993bbab40126f3800cb1daa3ee2c07e2bc87622cf818f3e6b1bf507", + "size": 7994 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/initial-docker-images.json": { + "sha256": "449d13f34e0b28d37629f60676696304188487bd8f7361f38ffae2ebbe0b4f60", + "size": 465 + }, + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/registration.json": { + "sha256": "03c10086892bbf06ed29011b8913dd30db7d4e87bb8131d61205b7f205ac59ae", + "size": 25368 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/config.json": { + "sha256": "b6060aba851f1c9ceb552f3dbe8af5eebcb2b41459271139e6cba12fb64a815a", + "size": 1792 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/job.log": { + "sha256": "72dbf76a42bac5c698689488a13aaa84a5443a9e95e0c0534c940e4b899a5397", + "size": 15166 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/lock.json": { + "sha256": "bf15c7ba6b68cae7db961380e8443a2eb62bcad83173adb4a463d090cb0fd79c", + "size": 5827 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/agent/bootstrap.json": { + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d", + "size": 19796 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/config.json": { + "sha256": "b1a77f7bfa50450bf69e99e4595bb54c986838a142ea3d4653dd1d33f3c2312c", + "size": 1598 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/lock.json": { + "sha256": "8b43a31c3f78e57457f848e653535c4161e11346bab5a7baedb919ee3b518e67", + "size": 2123 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/result.json": { + "sha256": "c3efdc8bb4fb1048f32aa4c88105b8c1a08262e98c2fbe329d41dd09853ae408", + "size": 4550 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/trial.log": { + "sha256": "c6f98f96c1eee7a4d8b3cfd2fda3a6603ead45c7c395196e8fa46d6fe1fa49f2", + "size": 7485 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/agent/bootstrap.json": { + "sha256": "a3c0057981c5d3db5eefa64571d2bcc3354657ae9602961506ab91df8197e94a", + "size": 94894 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/config.json": { + "sha256": "7c831d2d8998893cd68c2148b40b2cec6161d7a6bd3ac968ef4875c3e64a38c0", + "size": 1584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/lock.json": { + "sha256": "991a74f26a98961e10a1138c2b0a025d3dd8d8d9d61759b24182819766f71fda", + "size": 2116 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/result.json": { + "sha256": "1061d55f449577a08bfb866a4572bf1cda8efeba9e58e731b5d70a7972df0d22", + "size": 4508 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/trial.log": { + "sha256": "9e913a835b050cef2f099c026b81413b2f3b43243451a1b65f82fa5423e08caa", + "size": 7681 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/result.json": { + "sha256": "3741e74fa1ad5b99f0116f99d16fffe6c26762eee90ee1cbf5a9bbe25fd12f0c", + "size": 947 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/config.json": { + "sha256": "b17c74c68cfcab6fd9daea53b58e12192f8fd21de23051cc452c768043ac16f0", + "size": 3412 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/harbor.log": { + "sha256": "7cd9472480fab862e6f74b9b1b538ef206758b6477faf2266a07274d2d8f6cd2", + "size": 1057 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/manifest.json": { + "sha256": "0b6cdd290c51ed96c80243bf399bcaf1bc1698c17aa841b89fb94b6ee53d91a1", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "27bde1991d92f6becd6cb327eff9380e65ca4c4b771548f758c37b308223f052", + "size": 586 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "1590c3f38a3b2d17e64b8126a42b70b1cd381038e9a9c7b080c9d7bba9ea3d5e", + "size": 18705 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "dc582035cbe700e9081d98704c7b830f6d3b9b3065cf597b20e0111321fd4a67", + "size": 14993 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "4d051ee7f175e81f8a00096b1de47a01f1e75b1e4470bc3fe187e60f1b3e4e38", + "size": 15047 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "10d8856dad7b6c0e841f3e77be032eaf3e78a87d67e221b1386533cd989c451b", + "size": 18396 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "8b6421768909bd015a4cf2653969c7d70b3d190dcca2fdddb930efc067015eab", + "size": 14695 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/summary.json": { + "sha256": "57deb077e1d7b15b87d5cf898a1337e472ab3e87f8a796f230648e2467d692ac", + "size": 129304 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/config.json": { + "sha256": "c5a99fc325769009292fbc1c6a1ab0477c02b48425d773379579f5cb013e47a9", + "size": 1811 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/job.log": { + "sha256": "0fd993db1a1e6625d00f7ee85c329f345061a10319601aa52faa8fb635d3d9ac", + "size": 15166 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/lock.json": { + "sha256": "ba568cb0cbbfb3da596c31e95acd87f6ce58e22baf722bbc7f1e6faf52c7d192", + "size": 5879 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/agent/bootstrap.json": { + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d", + "size": 19796 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/config.json": { + "sha256": "1142928916e2a4553bf62f9842b23f3e316575f0c3805e8bc23ba0132333c80c", + "size": 1617 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/lock.json": { + "sha256": "c0f141c83433212847855e1e866909a329ad71aa033b126b951e3677f4046d6a", + "size": 2149 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/result.json": { + "sha256": "c2457b2afd8de269bfac9c2c67a0c120d1f2604bee58da87f3f01fef3bac4e37", + "size": 4575 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/trial.log": { + "sha256": "7cf8b72c79deabec9dc23ea8416f653592b928e266fdaaef81689df16ebdf9f6", + "size": 7485 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/agent/bootstrap.json": { + "sha256": "0d3b19888676a0f8c1bd7b850d8a349a49b07f1cebc2e1780e678948be935a60", + "size": 94894 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/config.json": { + "sha256": "09698f5d0538d5ee64726efe664ff2e00af4a7378c6461d61a56ac80bcda0a78", + "size": 1603 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/lock.json": { + "sha256": "06fbad5a6a4fedc1a506f673ecf32b8343cc19c7468c688cbc0569a3c3238b2f", + "size": 2142 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/result.json": { + "sha256": "942b30020eaf1fb96c31f6b9ff272eaf32364f08666f2ff3cb4388e4d089d486", + "size": 4533 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/trial.log": { + "sha256": "19cd5ac2957ba2e8653cb2f7246fda609a99c7e4cbb097018c1960bb1ba5d149", + "size": 7681 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/result.json": { + "sha256": "0c6f49e48dc0779321394b6ef60fe6ce0f1fbfc54bb856e12947106aedd34404", + "size": 947 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/config.json": { + "sha256": "380a9c937118753b46f82a603b853cb1dc7535bd87868aa57bc7199f0e7575a9", + "size": 3431 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/harbor.log": { + "sha256": "82d3adcae4de3db5017bae07df00040e02ca34bff92c0cd88b6c42cbc442052b", + "size": 1067 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/manifest.json": { + "sha256": "a1c0a37704126265f30150d7ab8d049adcbef97964b3ca57cc64c927dec3c50e", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "c612abd222e90c927f2c8ba2f933b5bbd8ab78d8e2e9b769d1f5975f1c890839", + "size": 592 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "623d6fa28119a45230454a693a1820e57ae0fb357758f3f8ea3d4d00a52af667", + "size": 18711 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "d2b29430a31f2bcbed44ec3b86815a890ba175f63ffcc061862ab25073996666", + "size": 14999 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "21d1dff276d39cfd86ff803e50dfb7e8d613c83d3c25f8cc948f868fa577a5cb", + "size": 15053 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "e1b494f9db2180b0dd9b02ad622da5459d6d10378923508158f626d53d53699a", + "size": 18402 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "6b62f2132ba2904920cd61f431f00dd16db424bbd1ba98bcf4a22e03013f7787", + "size": 14701 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/summary.json": { + "sha256": "7e60fc4d3c5a8ac06c48f630b8359cce05366334fd479d196c49e518f824ed56", + "size": 129304 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/config.json": { + "sha256": "2132e259cf77dd58e671031a3b89cf6cd912a3669d619c3af93cd77316e9f592", + "size": 1787 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/job.log": { + "sha256": "439563f59b244afebafb33c4a9fc129ad14ba90d056cfa20cfe14bcf1bb8b176", + "size": 15086 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/lock.json": { + "sha256": "d56d078e9755eda3e3da2224e0f6b3cd5f510914fa1d6341215c90284ac4a6f8", + "size": 5822 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/agent/bootstrap.json": { + "sha256": "569634e2774a66f6a021d30250a11667067e2fb7503586e9cffd6f8eadacdb75", + "size": 21911 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/config.json": { + "sha256": "2db85a02bc547eb41d58568cfc2d34ea41529585f0546eca2a03b43df74c0b70", + "size": 1594 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/exception.txt": { + "sha256": "a9418a49568b7803d4957ada7280037b57187c793a9cbed2016dd3d217b6afca", + "size": 5136 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/lock.json": { + "sha256": "0828a3fd0fde038023084f1081286134be902ed97903c7eba573d1604d4f6a2e", + "size": 2121 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/result.json": { + "sha256": "57b535c90e77e1bfb129bdb32e45f29e825132e0dfde847fcab2be97346db9b3", + "size": 10017 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/trial.log": { + "sha256": "6489443af7cd8e835f654b193dd6c780e8fb7f819db8aa7a3ed24bed8f68cb1b", + "size": 7563 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/agent/bootstrap.json": { + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826", + "size": 15230 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/config.json": { + "sha256": "2973f86c335ce2c79cebb9105d8c5736752aceb0a48bf886997d65bd7790cb44", + "size": 1578 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/lock.json": { + "sha256": "d64c3de3ad5d52896304f7f6d1fdd8f861ad029c8571e64e19ecaccf3efcd81f", + "size": 2113 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/result.json": { + "sha256": "bcddf59e8ca8e1042bc40c41d2928ef8a7ec7d419cc2fe234f42ae00ba304de2", + "size": 4490 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/trial.log": { + "sha256": "b2dca9a94a405af2eb206d9762d809739e383c2395b07cea68cccf9f6a5dc1b4", + "size": 7449 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/result.json": { + "sha256": "a87f912301b92fed28734e699ed5b7cbcfcfb38883fc9dca724246b8f7eb1948", + "size": 1092 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/config.json": { + "sha256": "927cf6c3cf9b6c1b7b770fd85c9238e700e476e5f9c2e9b80a093554213eacdd", + "size": 3407 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/harbor.log": { + "sha256": "2e13a641ebaef616252fce2d842da4974b9519913ea059773025139487971266", + "size": 1449 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/manifest.json": { + "sha256": "4edf18f054f462ca099b3c68eec1cdae8a9d6eeddb1ca486f3adb61c4d20275c", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "0aa0e443a1a8eb75ad6e5490c9729fcbfccdd594028796966cff3a07d25bf467", + "size": 586 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "5ea2f825cdce6b45397d71a75cb059b126600cc10aab18c9c8ec352e25a2d938", + "size": 18705 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "b49807a7592701fdec9c956f83403b200ae6b9f3a282dbfb70a1924d52d473b0", + "size": 14993 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "931dfcb17f33a8ae7f3361199fac1221083cc3c5f5aa4957bfcca5cc5c8f5db2", + "size": 15047 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "929b79aeac4bc486a71093da50983d18d0c5e768df5ea061559a144494409b08", + "size": 18396 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "f6fe9f941c0221f6c1f7c72b60294ab03372b09ab7083fdf7ce0a1fde52b839e", + "size": 14695 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/summary.json": { + "sha256": "087a66a22204170c7ac6d63ec1f52a59a8088f81c86bbf1e99d69f9f0201b4a4", + "size": 42333 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/config.json": { + "sha256": "86f5fdd95df5153cc97eb7d9d29583e29419191af55d8712beb4ade352da59e6", + "size": 1806 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/job.log": { + "sha256": "32ec9f9ef53e46e699dc05a8d920377bc2e816ab76fd88082857c046bbb1acf0", + "size": 14914 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/lock.json": { + "sha256": "ac0949e62ce9b8ee730cfdb09d32e1fd76692f9d8bd722fb6942580d5b78935a", + "size": 5874 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/agent/bootstrap.json": { + "sha256": "638922b88c59516844637223aff41c51866cf5e21824c22d7fd7116ebfb1e0fc", + "size": 21859 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/config.json": { + "sha256": "0b43d9685c51159f38c85a2539674c33fe03628adaef5fc23cb400c3f68366c2", + "size": 1613 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/lock.json": { + "sha256": "e1c137200ade8addc3425bb15921c07f28cad88dae614f0c4ffc3c95ed33dabe", + "size": 2147 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/result.json": { + "sha256": "ef17615f7c6a6c91d075791d90ad56adab7e245eb7bda5ce846bbfbf904f2366", + "size": 4563 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/trial.log": { + "sha256": "ffbf4c1e52a7b9f8b4deeb7810817f6c6e8fd66dfee71b7d616026ebe923933d", + "size": 7465 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/agent/bootstrap.json": { + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826", + "size": 15230 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/config.json": { + "sha256": "04574ff230cc8442fcd4bdead9a7d5f2c2d2f916555f61c3a84827ddef783be4", + "size": 1597 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/lock.json": { + "sha256": "8bff549930dd4d008f9e1dc07f9f46a6d79979df6be162e41c2aa3fe3b52aac5", + "size": 2139 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/result.json": { + "sha256": "f1e90e69169bf7cfee37a68f7915b981e4cff370caa457d9ba9dffaaa431b7cf", + "size": 4515 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/trial.log": { + "sha256": "bd3af473249fe27b80ea071c5062b212faf4c491e6c231d0763ae0d6753155ca", + "size": 7449 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/result.json": { + "sha256": "0329b3aff91d4535bb3b6b41ec69dca209140c6df67ddf5a31974e9a9322c849", + "size": 947 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/config.json": { + "sha256": "e83879378bc3dde4f386985bb0718ce54afa3bb8b2da3d33062f2f01110f6caa", + "size": 3426 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/harbor.log": { + "sha256": "404f2dd487d0d163a3cb7854c35743a51539a68f5cb687e7cbfbe0ccfba3a96a", + "size": 1068 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/manifest.json": { + "sha256": "bf7a0c43b592960af76176b8715b8440129886b6becc68f51dce7c2dafb06bcf", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "d3264411f2a8587d0f4ca96d9e63b2db363450dcd09bf9f14f8c3fcfd68d8b57", + "size": 592 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "9660c0bcff554a4b5962546562c7604fceeed961e1380adbd4de45be46f9c0ec", + "size": 18711 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "73304b7459ca56a62770223f9368732cce9aa96fe2e11a27f44bf347d12af88b", + "size": 14999 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "a9e107e0442fff3d3385662230d294ecb66232628002e83cdf4ce800c96c081e", + "size": 15053 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "3347796de01564e4c498a1f942655e531b2d2b764297ab97c931b4640f738b6d", + "size": 18402 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "628fa8cf06a8504bf9beaa39cdfa79043c9e9a67046716ef38ddc50b3d29fb19", + "size": 14701 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/summary.json": { + "sha256": "40582b6a4c73c6f9ad84f1e454e3072b5818e5b4213e08082f2b88a3b88f0aa1", + "size": 42255 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/config.json": { + "sha256": "f831ecad59eb127789fed1b1e74f251302123c556e887f13345efd90e1395bf4", + "size": 1820 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/job.log": { + "sha256": "740ed51ace3638207c365b5c387ca3373188eb7f8258961ee94153812d91b8af", + "size": 14954 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/lock.json": { + "sha256": "7537a6b89da461c16afa91bf74e9f458e743a17765d52be9afc0aa2730e46306", + "size": 5888 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/result.json": { + "sha256": "c7a91d2a57153465511c760ed27d7dc64a60c797eaaaedb7a8e7c82a000d37e7", + "size": 947 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/agent/bootstrap.json": { + "sha256": "a9460a42b2e140bbe0bee9d4a2226d95a34d556d4cdf92a6b310b8b3a72c3bca", + "size": 23414 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/config.json": { + "sha256": "b7cc691f9af2d590c3e149e509c578815474f2061207bc5f8d2a8433c8449e98", + "size": 1621 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/lock.json": { + "sha256": "b58959a6ca51865bfcd6fdd5efa23bfeb1e5166380f8166bce3347a88cc869fb", + "size": 2151 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/result.json": { + "sha256": "d29a589b7595b286187ff3dea2ec985151dfda8cb19da12fee759f75c59771bf", + "size": 4587 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/trial.log": { + "sha256": "cfd576cff8c053f2e9b1a26b35319e301d769fcdcd00d7ee37d7ffdeae937732", + "size": 7489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/agent/bootstrap.json": { + "sha256": "34af5d820280f7c297e211a7e512cc9101f09ab3908a0b7bf5ded9486eb61085", + "size": 23412 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/config.json": { + "sha256": "23d2ce7a5e40843c49819c29da25cd4c2a33fd0238744db36550922b9d8d9d58", + "size": 1617 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/lock.json": { + "sha256": "b50e0e078bdeb574d2135951fb3800c83bf99d2dda8f7cf80df3e90ca38c1a53", + "size": 2149 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/result.json": { + "sha256": "4862f8fa3a7b1038b169aa97a54f0892917ad45bdb769a3fb25fba62ade08979", + "size": 4575 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/trial.log": { + "sha256": "a25728b343f1989b1000a353b680b5411df575377194aaf0e442e14c28c67192", + "size": 7465 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/config.json": { + "sha256": "9d23650080334127301348fb0e301666a6c8e1eec7e72f5b5dc25db980ac6593", + "size": 3440 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/harbor.log": { + "sha256": "12efbc495e6d70f67b1ae4562a2eb13bb886eef64c629b67f1b315187fe4e5ee", + "size": 1068 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/manifest.json": { + "sha256": "a79d0cfc3f8c90c299dfdbf2f245115c51924cb56026116b53ccd54bc44abf9d", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "6f17f6d9bb1462f0246dfbdeecfe4a601f40ffc8b98493ec1436addcceaf0c3f", + "size": 592 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "b8b8a411958d0b502707cb70a53c1072bd44b0907846f0d5085be99603464f5f", + "size": 18711 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "247959e8cc2eb93f85e3f29c1159746f8f48c4195ec984a484daadb02bbd0e37", + "size": 14999 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "fc831988a61758b1f33805f9cf28863e02f314ae40a72e064e9855f7aa0659a4", + "size": 15053 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "539e713913e8467ac6ab56728be08d22dd26051ed5f0b7c4c87e5f0c03b09ec1", + "size": 18402 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "3a735206e43fb96e5c188fa6d03a4f4e0a9d27228115900786e5dc2b3e845957", + "size": 14701 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/summary.json": { + "sha256": "6ac52365e43f74afd60a3424faf8d05b7df928abeca72c1559362664ecd8b348", + "size": 53070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/config.json": { + "sha256": "3bcfedac04b5c4ac5a3e0bb3a98c1a713a2f3772c09c0c15760dd11ace87a0e6", + "size": 1599 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/job.log": { + "sha256": "35dc322211d97c40a3a8ba87258f8a3719d54dc5477dc19a68658dc91b9b4b31", + "size": 13397 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/lock.json": { + "sha256": "507e911aab64da458c1f583a0f5e702b865d480b79834ca6b95fe764d7250421", + "size": 3305 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/bootstrap.json": { + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d", + "size": 19796 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/.anonymous-user-id": { + "sha256": "2c3772fa069fb782be615b9664d1cad5fb667dc181522c0a898f909014fe6e18", + "size": 37 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "9617ac0778ce6d951153e9ea913ac1efd28fb43b14fc96a8fd0539668962ea42", + "size": 80650 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh-events.jsonl": { + "sha256": "560ffcbee9aa1d0d4a79d995bb14c03b86ec3c053c01ab5dff68bd460bb8f372", + "size": 89583 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh-result.json": { + "sha256": "6abfba7ae96e460386999b16f464b95dd2afdc5a6af20a183562e29213464a95", + "size": 520 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/config.json": { + "sha256": "d4fb29c2314b06c78c0f6e875822444accb7c74c5e47d2b130bd4d35ebd8dbec", + "size": 1575 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/lock.json": { + "sha256": "fee536be94d5a1647dd1a3216cd26ea2a153344b62a7190b57a0aa1883bb6a17", + "size": 2151 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/result.json": { + "sha256": "7964222816878953a237651d62e8035a0031f3d82e720fb8bd95e8bf0e266808", + "size": 34018 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/trial.log": { + "sha256": "35dc322211d97c40a3a8ba87258f8a3719d54dc5477dc19a68658dc91b9b4b31", + "size": 13397 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/attempts/1/ctrf.json": { + "sha256": "45569dd708aa3a09516a19263f9bff0309fe8187d6c6f8d8c5bd68bc0d54956e", + "size": 1462 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/attempts/1/test-stdout.txt": { + "sha256": "afd6eb483b54b284bd88c964f1cc2a21639044f192cd314df11374ea011a24a7", + "size": 1249 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/ctrf.json": { + "sha256": "45569dd708aa3a09516a19263f9bff0309fe8187d6c6f8d8c5bd68bc0d54956e", + "size": 1462 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/retry-summary.json": { + "sha256": "227986f8e26eeea0b1299ea126ca86aab0d0d7e9757aaa4b1ac76c6092bb1c20", + "size": 269 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/test-stdout.txt": { + "sha256": "afd6eb483b54b284bd88c964f1cc2a21639044f192cd314df11374ea011a24a7", + "size": 1249 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/result.json": { + "sha256": "893bebac0b12f33557d0aa93881af36fbbedb37eaca3d5506995f0ee02045f59", + "size": 1141 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/config.json": { + "sha256": "57d3f28a413f43f035a1308429d63405d5108808633e3aa0aa3411532e98e968", + "size": 3143 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/harbor.log": { + "sha256": "5870a6f29db039fbeb55f4faba70034df80a88d1d71927fc14a7f02d5fe06c25", + "size": 1302 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/manifest.json": { + "sha256": "45923d5f9fcb2414b4403b8176faaf171ea48aa38df654a35585db9d1987636b", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "db97d6ee3ae77b905f4e075cbf3fd57ebd554832d4c66f2f902f95c2ab3b64f8", + "size": 601 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "a2e97d01239f54f41f056a6d20dc98188bab291cf5be6ec78123f6fc53f1082c", + "size": 18720 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "e4b762644060b54011fbfce0b8ea4709f7fe576ed44d59b8d152e83b8e46e813", + "size": 15008 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "548f30a528fde968c9be4b0294f607f228bddae28b40dc508a5019b2db5328f2", + "size": 15062 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "e936ee8a90bfbdce845630631ea13d6cd78934554e42bc82a850c24f44c3fac5", + "size": 18411 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "b76a7cfb9e91507bd366b585faf23f6cbc64a76477aa8afa3a0d45fd4cbc9434", + "size": 14710 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/summary.json": { + "sha256": "b442d625a6d2bddf0f56e72b4cddc388a7063535139d886939c5c37a2006e753", + "size": 22967 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/config.json": { + "sha256": "843e5055a85eeebacca9fa2c21e660d22e6e33ac537f0d0518b24407ededc1c6", + "size": 1585 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/job.log": { + "sha256": "ccee04586e6eacb00d110fb180aa543f8849f6c61c336ffd4088bfbbbbe24dcf", + "size": 13593 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/lock.json": { + "sha256": "1a442e0987e940aca23a2f1f58290883bc8545f98eb576fc1b5ed7bcc1714bc1", + "size": 3298 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/bootstrap.json": { + "sha256": "0d3b19888676a0f8c1bd7b850d8a349a49b07f1cebc2e1780e678948be935a60", + "size": 94894 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/.anonymous-user-id": { + "sha256": "e6bb40b3ebccda5ba9c8a5f05720d61f9fa8d3e32fd8bd9d1c8930ecad0a57d3", + "size": 37 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/attachments/v1/objects/ac/acb96080e83322f5be24aa357055f8a178cac79cc82a23bae9d4f6113225ca51": { + "sha256": "acb96080e83322f5be24aa357055f8a178cac79cc82a23bae9d4f6113225ca51", + "size": 598 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/llm-deepseek/files-v3.json": { + "sha256": "81a910a63f4649c149af38289761d9464eab863f99a7779e19660f6a8323f4e4", + "size": 483 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "26c9f5ab32559672ce0acbbed914fde4e53ba478c5dd02d0a2b2663b872f0140", + "size": 427954 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh-events.jsonl": { + "sha256": "3f97af2928fc3eb8bb6cc8a053134dd123134d2ff7bc431b6fbdaa1582420f76", + "size": 488034 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh-result.json": { + "sha256": "7c4010a4f0456c1b6a8ee240e93ee2d108340af1e9d2f40866743dbe71ac9f1e", + "size": 354 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/config.json": { + "sha256": "cf7ad9124a6b97cd3edfb355ae62b0004792c78598adb69e7ad80e171db59ffb", + "size": 1554 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/lock.json": { + "sha256": "5fb9a320732aae3fdf9c717ab511dcccbbc41b90a26a7eb5fb6cf579d5e1fc17", + "size": 2144 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/result.json": { + "sha256": "52fb13759ec498d5fa96250033110210e7efcb9295d30475e6b37236059f996c", + "size": 140982 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/trial.log": { + "sha256": "ccee04586e6eacb00d110fb180aa543f8849f6c61c336ffd4088bfbbbbe24dcf", + "size": 13593 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/attempts/1/ctrf.json": { + "sha256": "84f668d0c8d2f53fbe53ac901c0c712e70a0aeed08189f7d6ee6374892007e09", + "size": 5268 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/attempts/1/test-stdout.txt": { + "sha256": "1776965440cf6137f1918f1aa073281347d52da046e477ced39824d828970356", + "size": 7732 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/ctrf.json": { + "sha256": "84f668d0c8d2f53fbe53ac901c0c712e70a0aeed08189f7d6ee6374892007e09", + "size": 5268 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/retry-summary.json": { + "sha256": "0a51e3b4774c831a3230ee69cbe7232f3300f7fd5c7794dc8e955a92ff0a9834", + "size": 269 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/test-stdout.txt": { + "sha256": "1776965440cf6137f1918f1aa073281347d52da046e477ced39824d828970356", + "size": 7732 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/result.json": { + "sha256": "389d66cf51653a39f5ad176422d0551e18c96930ff187813639d76d4a9d404cb", + "size": 1139 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/config.json": { + "sha256": "f8f2742c2ddab10e6d9ca235789ce2eb951428e14f44c1b44d722af3745da6a5", + "size": 3129 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/harbor.log": { + "sha256": "ab81954457f29c8e38797b63274602d32b920872bb37a0ee62ad3d6e4d2258f9", + "size": 1289 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/manifest.json": { + "sha256": "45923d5f9fcb2414b4403b8176faaf171ea48aa38df654a35585db9d1987636b", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "e43cdb58d72f38b8d0c9e07d8011bbb215260c185461d7027234dd8d957ba6e4", + "size": 594 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "8af2164960fb928073e8d880207898409f2108294ddba5a93c11ce7c8695a119", + "size": 18713 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "82f6bc4914351dcf90ff7b512e2e89e1a62727c470c2a2b29620695804a8ee09", + "size": 15001 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "b81ef5afeab875cb55fb9df3b08455bd74f058bc24fe7ab1302b0e5375d7f797", + "size": 15055 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "b101adaa535551f20fe56b3e8ca3e28c52e3e4a54576a51803f963ff7eee99ae", + "size": 18404 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "ce389e71828aeccc15e3b213c7da12ee0f698d34b24fef9bf14937f723bfec66", + "size": 14703 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/summary.json": { + "sha256": "c79325ee26f5cf52372922fc1790a49e5078f3eacfc66b015432a7ada428422b", + "size": 107111 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/config.json": { + "sha256": "b5bac91f76be9127e1ddfe57bc29596f85a1028669a9830cdf856e8ffcad3e22", + "size": 1595 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/job.log": { + "sha256": "8bab31f91eb2fb76a5bf25532af48033c0e11f967145470f4717564c54cf7291", + "size": 13377 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/lock.json": { + "sha256": "8caf35b6b4a2f839af6bdc81b832e4d75387acd81c15663a43b95802d7ac35d0", + "size": 3303 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/bootstrap.json": { + "sha256": "638922b88c59516844637223aff41c51866cf5e21824c22d7fd7116ebfb1e0fc", + "size": 21859 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/.anonymous-user-id": { + "sha256": "5f3ad2b9665279ce9f4a259b91c733c11eb9ace7c13ff5fcd764dc141eaf2cef", + "size": 37 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "d78912b555586297b5821012499be6bc693c85e549d90ff38eb8ddcd9972fccf", + "size": 273364 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh-events.jsonl": { + "sha256": "4db152a851b1538a1c1dc0dd28d6a89bce1131fad36cac4aa555b7f297d549b3", + "size": 305465 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh-result.json": { + "sha256": "5d60b1dae24a5f2217078efae2eb42f99fcc51ae4f235dd417c433e1d30e96ac", + "size": 676 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/config.json": { + "sha256": "53afdd0fd3a9932593423705418e446a8a4c1244789f842eec9c0ef72befd818", + "size": 1569 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/lock.json": { + "sha256": "4aff7e6eeaff6d59886ac6b466cee4dc716f0d6d9490bf614186874bf3773d31", + "size": 2149 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/result.json": { + "sha256": "1bd308679ac767e99b6814dfdc510d99ff51a9797a156ee8989bd8e94210075b", + "size": 36811 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/trial.log": { + "sha256": "8bab31f91eb2fb76a5bf25532af48033c0e11f967145470f4717564c54cf7291", + "size": 13377 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/attempts/1/ctrf.json": { + "sha256": "dc814bcfc80bc5a60a894700783f45f68cd77717d1da358b8b4278df129f6d28", + "size": 2138 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/attempts/1/test-stdout.txt": { + "sha256": "36929bebecc3c4ea17c0557f1a45ad5353768b62312508f366c1097fb91cd0d1", + "size": 2318 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/ctrf.json": { + "sha256": "dc814bcfc80bc5a60a894700783f45f68cd77717d1da358b8b4278df129f6d28", + "size": 2138 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/retry-summary.json": { + "sha256": "5f1d52539b277524eaaa8ed6879ef319d00882f26f696a1fc8a86f23bfa1811e", + "size": 269 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/test-stdout.txt": { + "sha256": "36929bebecc3c4ea17c0557f1a45ad5353768b62312508f366c1097fb91cd0d1", + "size": 2318 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/result.json": { + "sha256": "fe69f1b76fac910db93c4b56235e9df1c641016033eab292bf65405f20b82c0c", + "size": 1144 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/config.json": { + "sha256": "8cb93987ef5ecafed86ccf244582bf64dc7f8c62f121b0007482c7aeababd8f0", + "size": 3139 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/harbor.log": { + "sha256": "73d84f95acab3489dae73d166c1db04ad1a02c92228bc10a23ccd08ccc11aebb", + "size": 1298 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/manifest.json": { + "sha256": "4a239d7d2c769565b765d534fae50638d3bc49476237c72b4b6aa2ccbdacf20d", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "8519b27f526c3f66fbff55c0f9afcff6e8c53c3700e48c8a8cba6fef1ca86af2", + "size": 599 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "f867aaab3d96505f1d094d7ed46569866c81e97c425b77bc2fa2787442849fb0", + "size": 18718 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "ad34c7bfc9d242c2be36c2212c3a011c39e586c01d329bc339345cd8213ee14a", + "size": 15006 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "f61f2667dbbb16f3a6681bcb19c71a88e90b06f28c558d8cd60978c208becf29", + "size": 15060 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "84a4849bd1141190043f6a2b4fe7363292421a53c2d08ee7d9356fbc88be71d0", + "size": 18409 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "58b754604baafa3e8cdc526753bb09cab4936ce3ae5e9c069cb34d9987e392c4", + "size": 14708 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/summary.json": { + "sha256": "db40b4389894cf3ca72cc1751e61561db160d5ee50d2c214e41cdeea6e9f702a", + "size": 25236 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/config.json": { + "sha256": "ab9ff2fd05ca1b0c48642fb6f1b9ed5effc77882ab339512df7f0de2e5a99cd9", + "size": 1579 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/job.log": { + "sha256": "0b25533788e512b5c84baa5cd5e5af1d25817383b2459320c0fc46b947b403ff", + "size": 13361 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/lock.json": { + "sha256": "329e95e2d9385ab4cd18b8f3dcdf2febc08b5091fbb65e5529af101a163dc8fb", + "size": 3295 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/bootstrap.json": { + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826", + "size": 15230 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/.anonymous-user-id": { + "sha256": "7fa49c48f7d9d6628c304fff49a4a318047984c2d8be37c56e37f0ccabc095dd", + "size": 37 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "0b872374ac399d9a4ea96273882c0887d2e171ee201f08aa60532352cc49d892", + "size": 258580 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh-events.jsonl": { + "sha256": "5d116300a3fde02997933ca5c7c771956a34ad4dad749f0c4b5919009e3caddb", + "size": 293545 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh-result.json": { + "sha256": "7c9c256eff053d28ac305c585e384770389f4efe925e3696bceb7a582b17736d", + "size": 328 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/config.json": { + "sha256": "84ad50a5197042e38e0e571b0ca41eb41b8af6e7d7f0d2aed53686b0ebb4ac30", + "size": 1545 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/lock.json": { + "sha256": "f95d2b5f094f5a5d5b590d50b3be2bcbf6ddb16934004f28a1314be113a099f6", + "size": 2141 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/result.json": { + "sha256": "2a08fce9a87f184742e280a3004e3e7966296f5933ac72eabf2ec19764dc428b", + "size": 27306 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/trial.log": { + "sha256": "0b25533788e512b5c84baa5cd5e5af1d25817383b2459320c0fc46b947b403ff", + "size": 13361 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/attempts/1/ctrf.json": { + "sha256": "944d6d34bfe669f287a144833a4fd7d8efedb6cc9831b67dfe827c019b69d518", + "size": 2554 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/attempts/1/test-stdout.txt": { + "sha256": "5fdf69526b3cc8e8050cfe985df4b074b2ad9d569c5436eb4d3ddd3d2af9e572", + "size": 2238 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/ctrf.json": { + "sha256": "944d6d34bfe669f287a144833a4fd7d8efedb6cc9831b67dfe827c019b69d518", + "size": 2554 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/retry-summary.json": { + "sha256": "af00a7df4a76ece44d098028d1574868609704b24020c7a5f20670f4c8f9f953", + "size": 271 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/test-stdout.txt": { + "sha256": "5fdf69526b3cc8e8050cfe985df4b074b2ad9d569c5436eb4d3ddd3d2af9e572", + "size": 2238 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/result.json": { + "sha256": "e94dca53e294b261d2104d87ad01ce8052ab73284be37f59cda7b92dea9f2c15", + "size": 1136 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/config.json": { + "sha256": "dc9b57cc66a6b6139452383e8ed9fb740875394ae14f20cbd5cc2dc8fabeebc7", + "size": 3123 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/harbor.log": { + "sha256": "686a0592a5eca06d271a80a15f99c14715d55302743a2ef41af7f1b24eb924cd", + "size": 1282 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/manifest.json": { + "sha256": "970c1ccb0b7c324cee5eae9052eac48e3f15196465d48d61e947046cb2f6c2fd", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "ee5603c8c07ca8d988d46cd707952064483bacfddfa7911b95c32f35ab1159f9", + "size": 591 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "08d113e0db179c7e5296411b544c32517196aa3d161d75432c5c64cbbc3296f9", + "size": 18710 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "19e6966356202cd0f263cd83c45e21960960783a56541a140f13bff6c96f5f1c", + "size": 14998 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "9fc7728baf63fb9f853891765680ba0135c14f7e8f3cf99cbd4eb45878fc8bd8", + "size": 15052 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "a87225b6c19cf1483e99cf6128d5894eeaa97295c13d3f6c36ddf9aa2e34e6e8", + "size": 18401 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "457a04d4de7ad426216cc9f9df0a159444180c403212cad10740914bef68d8f8", + "size": 14700 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/summary.json": { + "sha256": "47cb7588ee4d19beff8f8f5bfb8f5b79dea0f186a36ffe16b6e7b1e22da5c3a4", + "size": 17795 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/config.json": { + "sha256": "ca35a0876af359667d88accfd4fb5b2ed653be708570068cc08fe9f2b1f09f3e", + "size": 1603 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/job.log": { + "sha256": "e34bb0e6981081cf6ca141cdba21fc4b8c7345e4b036cdd8be3c4036f3841092", + "size": 13401 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/lock.json": { + "sha256": "465d4eac6e50fed861e47eba488dd1a8e8079905e21165da7ad375fdb67e7087", + "size": 3307 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/result.json": { + "sha256": "e38b1cc5a6e596db91fbac55db6f3cd1974b02b0ce6abf999f9ddefe59ab492a", + "size": 1148 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/bootstrap.json": { + "sha256": "a9460a42b2e140bbe0bee9d4a2226d95a34d556d4cdf92a6b310b8b3a72c3bca", + "size": 23414 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/.anonymous-user-id": { + "sha256": "c5aa9aed82f7ef66f9898d1829509d3a42d9baec6ef48c27ccc76bc9d0df89e1", + "size": 37 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "cc8d36c82043c7304b9ba8533b8f1ed07b54bcdd5563a7c2018e29bb5015de25", + "size": 605403 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh-events.jsonl": { + "sha256": "388f8201b50db765149e4e93ec88cbc2e7814665e6c920a93c95df71b18b3fc6", + "size": 678797 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh-result.json": { + "sha256": "b840ab7c9a9b09dbb33d3d5f4b2e95941c8ee5ce287ab397ca4eec908990547f", + "size": 637 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/config.json": { + "sha256": "644af5c6a31e3f670c07ac1d0d393dd98adac321bb9ff53588ccfd21fdaa1c8a", + "size": 1581 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/lock.json": { + "sha256": "8f815de6eeae24e2bcebe43240cb8db615f7afe2d7e05f2402263ab8e9723623", + "size": 2153 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/result.json": { + "sha256": "7cb7906719d93c44ea780360553ecbf746b9d997a2f3c611b5c36f9d98c91f32", + "size": 38842 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/trial.log": { + "sha256": "e34bb0e6981081cf6ca141cdba21fc4b8c7345e4b036cdd8be3c4036f3841092", + "size": 13401 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/attempts/1/ctrf.json": { + "sha256": "7db34a6db947f1b4b8abbc682496408cc027e2ea15fa9c75820701552372d069", + "size": 8186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/attempts/1/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/attempts/1/test-stdout.txt": { + "sha256": "69cb6842eab805f1b2699abb5f51242c4a4fdc207b216668926abb7974645d04", + "size": 14491 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/ctrf.json": { + "sha256": "7db34a6db947f1b4b8abbc682496408cc027e2ea15fa9c75820701552372d069", + "size": 8186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/retry-summary.json": { + "sha256": "7443ea5e207b41cbcf0d965fe4ce34cba1e1aff258986ccd71d6e333754953fc", + "size": 270 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/reward.txt": { + "sha256": "9a271f2a916b0b6ee6cecb2426f0b3206ef074578be55d9bc94f6f3fe3ab86aa", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/test-stdout.txt": { + "sha256": "69cb6842eab805f1b2699abb5f51242c4a4fdc207b216668926abb7974645d04", + "size": 14491 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/config.json": { + "sha256": "2e47fe00a10c5ceeb0ab4d649e17751289dd24ed4654f4f212a1874a5fdde23d", + "size": 3147 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/harbor.log": { + "sha256": "1cbbb8139b524dbe7d58ec0f7847947c2ce7568ab33cfaa8c7aaf81783193a22", + "size": 1307 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/manifest.json": { + "sha256": "986021e2750ffa1f29629d8ef3188b555535460ab5ea7769af0e6d7bb6bf0a69", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "58a80c9e7218804d4dcdb8467ce9b8da78e0f1b01e0aea2407da801195b517f6", + "size": 603 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "daa8b0737d9b003f044c0ca5574da397c097faf5632ec868ce7e49f050a0fe2a", + "size": 18722 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "544183222a2119560e1f4a4e5edf84f763732c6b3b0a34b0bf67a0ec2a218e3c", + "size": 15010 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "4233c471bd86bb842a91087b68832887bec1ea70ac74c9a20e2668e8084c6783", + "size": 15064 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "0880bb9f947f892dbca4bb82a5852af511efc51a2144f2f36e1b475d5b1c3c7b", + "size": 18413 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "a792a4063ffb6eb1b9da27c47d97302fde93d6308f1cb88776d09731f37d8d02", + "size": 14712 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/summary.json": { + "sha256": "497652c2f564b59becf49f9f1533665553850318ba08a7a37026528e51a76c06", + "size": 26926 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/config.json": { + "sha256": "964f95d7c3503cb623b54e5d670d98b813d46377ef0fbba7d5cc7c23a1c9dac8", + "size": 1599 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/job.log": { + "sha256": "1f23853f873c0fcf3b8ad51c194c4f451fbcb57281a585138215459b238dfdaf", + "size": 13377 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/lock.json": { + "sha256": "d2e3c8f43cf14a5ce2aa7abcf392f8b51c780c5e665682920faa320cdf8a4295", + "size": 3305 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/result.json": { + "sha256": "3bfff2e1ed77c2df1e1745d2604efb6c2c67a5cf16a36e851610ec59a26722ed", + "size": 1143 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/bootstrap.json": { + "sha256": "34af5d820280f7c297e211a7e512cc9101f09ab3908a0b7bf5ded9486eb61085", + "size": 23412 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/.anonymous-user-id": { + "sha256": "e74a76edfb9bdb2b3ecd5d4f3670bdc7ad7d2427e9cb93fca04db379131a3194", + "size": 37 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/profiles/sdk-minimal/cordis.patch.yml": { + "sha256": "ef189a8c27db6d63930aa3046a3040482e952eafcb7487c644d508e8d461f027", + "size": 217 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/profiles/sdk-minimal/cordis.yml": { + "sha256": "c300dcf2ebc5f02062d6591268d29d3db6fe45e0cb138f5467276fe2ba06076e", + "size": 223 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/profiles/sdk-minimal/package.json": { + "sha256": "73c089627acc47000fd21b1a47445f166241b867cfdd59eb1eb66b269810ed93", + "size": 186 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/profiles/sdk-minimal/pnpm-workspace.yaml": { + "sha256": "ae7c5b68e2f157528e62885804e69e88583897b775e03c86fcbe52feaf498aba", + "size": 61 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/sessions/--app--/benchmark/session.lock": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh/sessions/--app--/benchmark/session.v4.jsonl": { + "sha256": "06e62c092e90cda5d1c0bb1e5acdd683651b68f46d2c07f005799e14d3511d54", + "size": 222423 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh-events.jsonl": { + "sha256": "d1f6f569271f12e30e9133598d27d0586229879675c85aba01ce729f87c9bfa8", + "size": 246189 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh-result.json": { + "sha256": "244ad977b56376e5079acdf82ce1403d82c2a7b7a58ef04c9def8537a15f07e3", + "size": 678 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh-stderr.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh-stdout.txt": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/process-supervision-required": { + "sha256": "d70211a2593233914074a8fc68d4a539a3512d072ed544e169c7d734e9bfbc9b", + "size": 17 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/process-supervision.json": { + "sha256": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "size": 42 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/artifacts/manifest.json": { + "sha256": "c1db653f487fb773c4f9247162ab3314e7f0584a486914b9874c4b3d6949e42d", + "size": 159 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/config.json": { + "sha256": "73e6e8356a969133cb6e6bb6831078bd483ac566438f92e19b464b83b53b5486", + "size": 1575 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/lock.json": { + "sha256": "602009b5a8adc30407e66d4417004d0ac2ba8dc4b840d35185fd36d0ec5cfe90", + "size": 2151 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/result.json": { + "sha256": "c4b2c2af77c402de31017564e29740814d413752b6d306253fa90a29b6bc5360", + "size": 38820 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/trial.log": { + "sha256": "1f23853f873c0fcf3b8ad51c194c4f451fbcb57281a585138215459b238dfdaf", + "size": 13377 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/attempts/1/ctrf.json": { + "sha256": "17ece9a78ebfb0f09dc4c9be99deb5c46ccb890b24f19d01823c36e22be45713", + "size": 1772 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/attempts/1/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/attempts/1/test-stdout.txt": { + "sha256": "92f8c999585b93e3575786304e330d327b043e24eec9b2080df45617ea782b88", + "size": 13635 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/ctrf.json": { + "sha256": "17ece9a78ebfb0f09dc4c9be99deb5c46ccb890b24f19d01823c36e22be45713", + "size": 1772 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/retry-summary.json": { + "sha256": "2d71d0cb1ba3c6a823496a68f073fc3b0806c73b323867d1e059d2c2c7b615e3", + "size": 270 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/reward.txt": { + "sha256": "4355a46b19d348dc2f57c046f8ef63d4538ebb936000f3c9ee954a27460dd865", + "size": 2 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/test-stdout.txt": { + "sha256": "92f8c999585b93e3575786304e330d327b043e24eec9b2080df45617ea782b88", + "size": 13635 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/config.json": { + "sha256": "419a58bbc8b1b219ebebf279ee1d5fc0f253880d89c3ec6d45f2cdbf5ce71b91", + "size": 3143 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/harbor.log": { + "sha256": "65f3979eb0440cfe3968b97fbbcd3fc82b5f5c1796996b0317ec3d0e89db71d8", + "size": 1302 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/manifest.json": { + "sha256": "a0a307644bff0bc52ff2e0aa10d2f2cbc20ed9932cf5ec610d1a1992ec34b35a", + "size": 3669 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__init__.py": { + "sha256": "d25ed286e8a9f9f78ca9980ef041e12b4a12d0a274ab60984c633fd7e2e5ef1a", + "size": 315 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__main__.py": { + "sha256": "e322e82a6b7ac37a9dc3429c2752ad4d212a82acefa4f2bc0072b2c87e82afa4", + "size": 15679 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/__init__.cpython-313.pyc": { + "sha256": "6a8dcc44fd2094ed622c48314a9108b24615da540ebfa247c218e615061473c8", + "size": 601 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/adapter.cpython-313.pyc": { + "sha256": "5a9e017bf4ec0fe85faa9aa02121fd8149c6b66b2e9167adf3f900d9d1ac7651", + "size": 18720 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/bootstrap.cpython-313.pyc": { + "sha256": "d65fa6905e38524fb27ecc07e2fe2dac1a08f846a369aa40ab796174366ef30a", + "size": 15008 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/deepseek.cpython-313.pyc": { + "sha256": "9d01eafd2ad29fbc6d31dec5b6c4ef85fb7b9ae4bbb521efd604c366c75acd1a", + "size": 15062 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/pi.cpython-313.pyc": { + "sha256": "ba476fcab4ca0700ac5cfa74fd834b9e4198ba08ab4abd44a4c56a60e4d78bb1", + "size": 18411 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/__pycache__/verifier.cpython-313.pyc": { + "sha256": "bfff4d871b867762deec172e6607b48fec803905abdbb23f0e21257e71ac752c", + "size": 14710 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/_agent_runner.py": { + "sha256": "171c880501bdfc7ea4e85157daa6f8d3b48b54d66e10a7e97f292b6bc2e34cd4", + "size": 5489 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/_dsh_client.py": { + "sha256": "4c49c2bc682993044117ac36be0b17164c6f1122e4598f4245c21afc061d6780", + "size": 2140 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/_verifier_runner.py": { + "sha256": "1733a2681bd9d13a30fcc1ab5317df1025d28d2d082b8cc9de85c0d9d95e723c", + "size": 11487 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/adapter.py": { + "sha256": "d1aa154ec6972d706f0725cfa69c56635506d4499dff2a129e2c957ea7034c02", + "size": 13070 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/bootstrap.py": { + "sha256": "5d7f086e5eb92301e4e43f5473b8f4bcdc012f6950bc65574ee3ff07d66ef7b8", + "size": 11246 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/deepseek.py": { + "sha256": "e76e130701744a02f26ccdc3da5b41ac98c491044025bd506b5fd52901b55b86", + "size": 9697 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/pi.py": { + "sha256": "e703174d76a1b670466ba06904db6b08a26769d42ac974274f61423ff0778fd6", + "size": 12355 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/retained.py": { + "sha256": "2f1bc92128ade805e9fa6cbf02bb8aae3da60486423785f422152f837b57e36d", + "size": 10584 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/source/harbor_adapter/verifier.py": { + "sha256": "772c036a60002ee140ce37a87f7a859923ee307d9ad229f4349e2d44be127767", + "size": 10016 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/summary.json": { + "sha256": "aa9f8657a890c4e8e6f8ba10efb4a7c5d2b330ef75ec24ac105314743582c4d0", + "size": 26920 + }, + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/working-tree.patch": { + "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", + "size": 0 + }, + "jobs/dsh-freeze-supplement.py": { + "sha256": "f7223219adf31a2fd2eb0f60d7d2e9e35b1c9c6ca1639fbae8881368e2307878", + "size": 3513 + }, + "jobs/dsh-probe-pypi-transport.py": { + "sha256": "02d43b929e97cb8a73c32525203cb9fa439294aa3dbfb55e804a41c009483237", + "size": 2294 + }, + "jobs/dsh-pypi-range-download-probe.json": { + "sha256": "307b62f3b83be939d7e43c47a4bc5bb8cbc2f029a4378a8450041cfa036f8b06", + "size": 522 + }, + "jobs/dsh-pypi-torch-probe-metadata.json": { + "sha256": "5475befa5167bea5dbd6862780d8f769f1276b34e7836136c572ad2a009e6619", + "size": 495 + }, + "jobs/dsh-pypi-uv-smoke-audit-events.jsonl": { + "sha256": "330958a0b95493071fc429c43ed8c4f25a96ff648b88ff4531dc29fd9976b796", + "size": 6660 + }, + "jobs/dsh-pypi-uv-smoke-audit.json": { + "sha256": "19330565cbb59f9905b92df2c897d05ce1dda712ec02e516cf3e12ce3293ca98", + "size": 930 + }, + "jobs/dsh-pypi-uv-smoke.json": { + "sha256": "34c3adb735a8b973caa6d2b068ba618f92cd5888bdda1c8d1c3debea1518bfcb", + "size": 648 + }, + "jobs/dsh-supplement-final-validation.json": { + "sha256": "e76bd2160a51e26c8204eb4a875a227f759d1bfb6437090b5e73e8b0e3738fe3", + "size": 579 + }, + "jobs/dsh-supplement-model-v2-20261101.log": { + "sha256": "7f0dafcd730fa8003e8c78c1f5c92263d2c42dee9b49be38943d8605a27a1e7d", + "size": 630 + }, + "jobs/dsh-supplement-preflight-20261101.log": { + "sha256": "229c3db700d0031d22ca614f0245f3aeeac5a50c4d9224e134e34848ecfea8b0", + "size": 1954 + }, + "jobs/dsh-supplement-preflight-boundary-stop.json": { + "sha256": "1a058982408952541e9a98968d4c7e85f36ba050fe6693b1ef994b08ab3b5800", + "size": 424 + }, + "jobs/dsh-supplement-preflight-v2-20261101.log": { + "sha256": "02d9506ebfef29c4c94261fb6e31ae2259f569ea20d1afb84c161c44fc8feec3", + "size": 329 + }, + "jobs/dsh-validate-supplement.py": { + "sha256": "9e43ec631502d189a42ac7c39fe0d533a0632ec1ebc78ec1844188195ed9d439", + "size": 3167 + } + } +} diff --git a/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md b/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md index 72f0a3c..91b5c8b 100644 --- a/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md +++ b/benchmarks/harbor/reports/en/deepseek-harness-dev34-20261030.md @@ -1,5 +1,9 @@ # Official DeepSeek Harness on Pi's 34-task development pool +Follow-up: the separately authorized [November 1 six-task supplement](deepseek-harness-dev6-supplement-20261101.md) +obtained four passes and two scored failures on these six unscored tasks. The +original results and audit below remain unchanged. + All **34 registered jobs finished on October 31, 2026**: **22 official passes, six scored failures, and six unscored errors**. There were 32 model executions; two jobs failed during setup. Two scored failures have a confirmed SDK service diff --git a/benchmarks/harbor/reports/en/deepseek-harness-dev6-supplement-20261101.md b/benchmarks/harbor/reports/en/deepseek-harness-dev6-supplement-20261101.md new file mode 100644 index 0000000..3065842 --- /dev/null +++ b/benchmarks/harbor/reports/en/deepseek-harness-dev6-supplement-20261101.md @@ -0,0 +1,161 @@ +# Six-task supplement for the official DeepSeek Harness + +The six previously unscored tasks were rerun on **November 1, 2026** after the +maintainer requested fresh attempts. The supplement produced **four passes, two scored failures, and zero unscored errors**. +All six model executions completed normally, and every official verifier finished on its first attempt. + +The original [34-task experiment](deepseek-harness-dev34-20261030.md) remains +**22 passes, six failures, and six unscored errors**. These six attempts are a +separate, outcome-selected development supplement. Using the 28 original accepted attempts plus these six new attempts gives descriptive mixed coverage of **26 passes, eight failures, and no remaining unscored tasks across 34 tasks**. They are not an +independent benchmark or a best-of-two score, and do not establish a causal +harness improvement. The original six scored failures were not rerun. + +The [audited results](../../results/deepseek-harness-dev6-supplement-20261101.json) +retain every new trial, native reward/exception, analytical outcome, official +test count, timing, usage, and evidence hash. The +[v2 registration](../../configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json) +and [configuration history](../../configs/deepseek-harness-dev6-supplement-20261101/README.md) +record the changes made before model work. All six new outputs were +[frozen](../../configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json) +at **2026-11-01 16:56:07.773103 UTC**, before reviewing new trajectories or hidden assertions; +**590 files** were rehashed. Raw logs remain local. + +## Per-task results + +All six had analytical score `null` in the original experiment. `—` denotes an +unscored outcome or no model execution. Pi values retain its historical audit; +`‡` marks its pipeline pass with public reference-test/solution exposure. + +| Task | Original unscored cause | Pi audited | New raw | New audited | Official passed / failed | Agent seconds | +| --- | --- | ---: | ---: | --- | ---: | ---: | +| multi-source-data-merger | Verifier: pyarrow download failed | 1 | 1 | Passed | 3 / 0 | 29.6 | +| pytorch-model-cli | Setup: uv installer TLS error | 0 | 0 | Failed | 5 / 1 | 628.4 | +| pytorch-model-recovery | Setup: 1,800-second timeout | 1 | 1 | Passed | 5 / 0 | 110.1 | +| query-optimize | Execution: terminal API transport error | 0 | 1 | Passed | 6 / 0 | 270.5 | +| torch-pipeline-parallelism ‡ | Verifier: CUDA dependency download failed | 1 | 0 | Failed | 2 / 2 | 632.4 | +| torch-tensor-parallelism | Verifier: timeout, then torch download failed | 1 | 1 | Passed | 13 / 0 | 73.4 | + +- **pytorch-model-cli**: Completed normally. Five official checks passed; test_cli_tool_output failed because 14 of 50 tested MNIST predictions differed from the fixed expected values (image 0: 7, expected 2). Artifacts and executable checks passed. This is a scored assertion failure, not a dependency or verifier error. +- **torch-pipeline-parallelism**: Completed normally. Existence and no-hooks checks passed; both world-size cases failed backward-value comparisons at microbatch 0. World size 1: rank 0 lm_head.bwd max difference 0.04166729375720024. World size 2: rank 0 model.layers.1.bwd max difference 0.01778167486190796. The worker exceptions carry these official assertions, not verifier infrastructure errors. + +These are official findings, not exhaustive root-cause proofs. + +The six accepted evaluations ran **37 official cases: 34 passed, three failed, zero errors/skips**. Agent self-tests are not included. Native rewards and standard summaries are preserved; an official +zero is not automatically a valid scored failure. Setup errors, terminal API +errors, and verifier dependency/errors stay separate. A native agent timeout +can be scored only when its stop boundary is confirmed and official tests +complete successfully as an evaluation. Here, all **six completion boundaries** were confirmed, with **zero events after agent end or at/after verifier start**, **zero agent timeouts**, and **zero extra verifier attempts**. + +## Preparation failures and verified transport + +The initial direct-download preparation used `UV_HTTP_TIMEOUT=120`. Three +installation-only trials passed: data merger, PyTorch CLI, and query +optimization. Model recovery reached its **1,800-second setup timeout** in +verifier dependencies. The controller stopped after that completed second pair, +before the final pair or any model job. No active trial was interrupted. The +[initial evidence](../../configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json) +retains all four actual trials. Across the six selected tasks: **three passed, +one setup failure, two not started, zero model executions**. The initial model queue was superseded, never run. + +A reusable optional PyPI wheel sidecar was added before v2 registration. It +forwards current original PyPI index metadata while preserving versions, +requirements, yanks, hashes, and extra-index settings. A wheel is served only +after matching current original release filename, size, SHA256, and HTTPS +origin and verifying the complete downloaded bytes. Cross-process locks protect +shared downloads; unavailable metadata and corrupt cache entries fail closed. +Source distributions and original extra indexes bypass this wheel cache. + +V2 sets `UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple` and +`UV_HTTP_TIMEOUT=600`, allowing a cold wheel to finish verification before any +body is served. Setup stays **1,800 seconds**. Original APT sources and verified +APT-index/uv/Python-release transports remain unchanged. No task image, package +pin, official test, or native agent/verifier limit was changed. Package caches +and new dependency preparation can affect runtime; this is not an isolated test +of model behavior or download transport. + +All **six fresh v2 installation-only checks passed before any model job**, +including SDK initialization, Linux supervision, and exact-revision dependency +profiles. Five profiles were added; model recovery retains its existing one. +They preserve original Python/package/index requirements and run only +`pytest --version` from `/tmp`, without executing task code or assertions. +[Preflight evidence](../../configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json) +and [launch registration](../../configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json) +precede model work. All six formal jobs repeated and passed the same dependency profiles. CLI setup took **1,690.4 seconds**, within the unchanged 1,800-second limit; a passed earlier preflight did not eliminate later download latency. + +The full Harbor suite passed **301 tests**, including **16 wheel-cache contracts**. +A real original **uv 0.9.5** transport smoke with a fresh temporary uv cache also +passed, with **zero model calls**; [probe records](../../configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json) +preserve both smoke versions and a verified range-download experiment. +The frozen transport log contains **675 original release-metadata events and 270 verified wheel responses: 224 cache hits, 46 cold responses, and 53 unique artifacts**, across **12 sidecars**, with **zero recorded transport errors**. Every wheel response has matching earlier original-release metadata in that sidecar. + +## Preserved model configuration and audit + +| Property | Value | +| --- | --- | +| Upstream | `deepseek-ai/deepseek-harness` `0.2.2-alpha.2`, source `f3803c96974ad4a94195fb97e5c637e9c2a3afe3` | +| Profile | `sdk-minimal`; same offline SDK/runtime distribution as the original run | +| Model/API | `deepseek/deepseek-flash`, high reasoning, `https://api.deepseek.com/anthropic` | +| Generation | Native prompt/tools/compaction/request retries; temperature and SDK max_tokens unset; native ceiling 256,000 | +| Adapter | Shared frozen source `eff073b1601435305155cad92b6ee9e2250fbbdc` | +| Transport source | `dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3`; registered file hashes checked | +| Limits | Agent 900 seconds each; verifier 900 seconds, except query 1,800; setup 1,800 | +| Scheduling | Original relative order; concurrency two with pair barriers; one new formal attempt per task, no automatic whole-trial retry | +| Verifier retry | Only timeouts, at most three total attempts on unchanged output and original per-attempt limit | + +The only frozen adapter change from the original source is reviewed dependency +preflight in `bootstrap.py`; the model driver, prompt integration, supervisor, +and verifier are byte-identical. The user's upstream checkout remains untouched. +Fresh isolated sessions receive no previous output, diagnosis, solution, or +personal configuration. The remaining **55 tasks stay unopened**. No SDK +service-lifecycle fix is included; the original KV/PyPI failures remain intact. + +The official [agent integration guide](https://api-docs.deepseek.com/quick_start/agent_integrations/) +checked November 1 still maps the alias to DeepSeek-V4.1-Flash-1017, as recorded +in the [alias evidence](../../configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json). +It is mutable and does not prove the same backend checkpoint across dates. +Pi used OpenAI-compatible transport and a 384,000-token ceiling. Its sampler/WAL +scores have invalid/unverified legacy boundaries, and its pipeline pass has the +reference-exposure qualification. On this selected six-task subset, Harness and Pi each passed **four**; Harness gained query optimization and lost pipeline parallelism, with the Pi exposure qualification retained. On the **32 common accepted tasks** in the mixed coverage, Harness passed **25**, Pi **26**. Gains are DNA assembly and query optimization; losses are KV, PyPI, and pipeline parallelism. The KV/PyPI lifecycle and Pi pipeline exposure caveats apply to those losses. + +After freezing, **98 recorded tool calls** were scanned and **three candidates with paired outputs** reviewed: two CLI calls accessed public MNIST data, and one pipeline call inspected the root/Python paths and environment. No benchmark reference tests/solution were observed in those records. The inspected environment contained no sensitive-named credential entries, and exact matching found the authorized provider key in **zero of 590 frozen files**. This targeted audit is not an exhaustive proof of no exposure or +exfiltration. Known SDK usage is **1,298,682 input tokens**, including **1,242,112 cached tokens**, and **79,429 output tokens**. All six trials report `usage_is_partial=false`; cost remains unknown. Cache tokens are included in input, not added again; +these are SDK usage subtotals, and no provider bill was queried. +The formal queue ran from **2026-11-01 15:50:21.232772 UTC** to **16:55:21.072680 UTC**: **3,899.8 seconds** of wall time and **1,744.3 summed agent seconds**. Queue time includes setup, verification and pair barriers, but excludes earlier preflights and transport probes. + +```text +v2 plan SHA256: 69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93 +freeze SHA256: ee6a3c6a4f57b6da2ffcc04e81e54e888949285054aff8e7a218ce0f330671f3 +results SHA256: 412a0d0acd266d5573ff722a2e5376ab0264b457821d0d603cf8b3f0e663d067 +preflight evidence SHA256: e973e6d280171287c8c60b7888834b56259f897aac67e8ffa6c23daad39e3e06 +original results SHA256: e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396 +source archive SHA256: 0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4 +distribution SHA256: d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6 +runtime SHA256: 28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2 +wheel-cache source SHA256: ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74 +wheel audit SHA256: ef8385be4d1387b72014dca1e89b1281fed347714f9bd3e5f9e050b501e7366a +``` + +## Reproduction + +Use the shared frozen adapter checkout and registered local artifacts. The +orchestrator invokes the standard `python -m harbor_adapter run --config ...` +entry point; it does not create per-job adapter copies. The published configs +retain local absolute paths, so reproduce with equivalent pinned artifacts and +new job names. Do not overwrite or replay the registered jobs. + +```bash +export UV_PROJECT_ENVIRONMENT="$PWD/jobs/dsh-harbor-venv" +export UV_CACHE_DIR="$PWD/.cache/uv" +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json \ + --phase preflight +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json \ + --phase model +``` + +Supply the authorized provider key only through the process environment. +English and Simplified Chinese reports contain the same results, caveats, +tables, identifiers, commands, and hashes. diff --git a/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md b/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md index 6c8590d..502d15e 100644 --- a/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md +++ b/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev34-20261030.md @@ -1,5 +1,8 @@ # DeepSeek 官方 Harness 在 Pi 的 34 道开发题上的结果 +后续:另行授权的 [11 月 1 日六题补测](deepseek-harness-dev6-supplement-20261101.md) +在这六道未计分题上获得四题通过、两题计分失败。以下原始结果和审计保持不变。 + **34 个注册作业已于 2026 年 10 月 31 日全部结束**:**22 道官方验证通过、 6 道计分失败、6 道因错误不计分**。其中实际执行模型的有 32 道,另两道在安装 阶段失败。两道计分失败存在已确认的 SDK 服务生命周期限制。本报告描述通过 diff --git a/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev6-supplement-20261101.md b/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev6-supplement-20261101.md new file mode 100644 index 0000000..3689475 --- /dev/null +++ b/benchmarks/harbor/reports/zh-CN/deepseek-harness-dev6-supplement-20261101.md @@ -0,0 +1,142 @@ +# DeepSeek 官方 Harness 的六题补测 + +维护者要求重新尝试后,之前未计分的六道题于 **2026 年 11 月 1 日**补跑。 +本次补测得到 **4 题通过、2 题计分失败、0 题未计分错误**。六次模型执行均正常完成,每个官方验证器均在第一次尝试内完成。 + +原始 [34 题实验](deepseek-harness-dev34-20261030.md)仍为 +**22 题通过、6 题失败、6 题未计分错误**。这六次尝试是独立记录、根据原始结果 +选取的开发补测。将原始 28 次有效尝试与这六次新尝试结合,得到描述性的混合覆盖结果:**34 题中 26 题通过、8 题失败,无剩余未计分题目**。 它们不构成独立基准或两次取优的成绩,也不能证明 +harness 的因果改进。原始已计分的六道失败题没有重跑。 + +[审计结果](../../results/deepseek-harness-dev6-supplement-20261101.json)保留每个 +新试次、原始奖励/异常、分析结果、官方测试数量、时间、用量及证据哈希。 +[v2 登记](../../configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json) +和[配置历史](../../configs/deepseek-harness-dev6-supplement-20261101/README.md) +记录模型运行前的调整。全部六道新输出于 **2026-11-01 16:56:07.773103 UTC** +[冻结](../../configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json), +之后才查看新轨迹或隐藏断言;重新校验了 **590 个文件**。 +原始日志保留在本地。 + +## 逐题结果 + +六道题在原始实验中的分析分数均为 `null`。`—` 表示未计分或没有模型执行。 +Pi 数值保留历史审计口径;`‡` 标记其接触了公开参考测试/解答的流水线并行通过结果。 + +| 任务 | 原始未计分原因 | Pi 审计分数 | 新原始奖励 | 新审计结果 | 官方通过 / 失败 | Agent 秒数 | +| --- | --- | ---: | ---: | --- | ---: | ---: | +| multi-source-data-merger | 验证:pyarrow 下载失败 | 1 | 1 | 通过 | 3 / 0 | 29.6 | +| pytorch-model-cli | 安装:uv 安装器 TLS 错误 | 0 | 0 | 失败 | 5 / 1 | 628.4 | +| pytorch-model-recovery | 安装:1,800 秒超时 | 1 | 1 | 通过 | 5 / 0 | 110.1 | +| query-optimize | 执行:终止运行的 API 传输错误 | 0 | 1 | 通过 | 6 / 0 | 270.5 | +| torch-pipeline-parallelism ‡ | 验证:CUDA 依赖下载失败 | 1 | 0 | 失败 | 2 / 2 | 632.4 | +| torch-tensor-parallelism | 验证:超时后 torch 下载失败 | 1 | 1 | 通过 | 13 / 0 | 73.4 | + +- **pytorch-model-cli**: 正常完成。五项官方检查通过;test_cli_tool_output 失败,50 个测试 MNIST 预测中有 14 个与固定期望值不符(图像 0:预测 7,期望 2)。产物和可执行性检查通过。这是可计分的断言失败,不是依赖或验证器错误。 +- **torch-pipeline-parallelism**: 正常完成。存在性和无 hook 检查通过;两个 world size 用例均在 microbatch 0 的反向值比较中失败。world size 1:rank 0 的 lm_head.bwd 最大差异为 0.04166729375720024。world size 2:rank 0 的 model.layers.1.bwd 最大差异为 0.01778167486190796。子进程异常来自这些官方断言,不是验证器基础设施错误。 + +这些是官方测试发现,不是穷尽性的根因证明。 + +六次有效评测共执行 **37 个官方用例:34 个通过、3 个失败、0 个错误/跳过**。不计入 agent 自测。 原始奖励和标准汇总保持不变;官方零分不自动等于有效的计分失败。 +安装错误、终止运行的 API 错误和验证器依赖/执行错误单独列出。原生 agent 超时 +只有在停止边界已确认且官方测试完整执行、形成有效评测时才可计分。本次 **6 个正常结束边界**全部确认;**agent 结束后或验证开始时及之后的事件数为 0,agent 超时为 0,额外验证尝试为 0**。 + +## 准备失败及校验下载 + +最初直接下载的准备方案设置 `UV_HTTP_TIMEOUT=120`。数据合并、PyTorch CLI 和 +查询优化的三次仅安装试次通过。模型恢复题在验证器依赖阶段达到 +**1,800 秒安装时限**。控制器在第二组完整结束后停止,最后一组和所有模型任务 +均未启动,没有中断任何运行中的试次。 +[初始证据](../../configs/deepseek-harness-dev6-supplement-20261101/preflight-v1-evidence.json) +保留全部四次实际试次。六道选中题目的状态为:**3 次通过、1 次安装失败、2 次未启动、0 次模型执行**。 +最初的模型队列被后续方案取代,从未运行。 + +v2 登记前增加了可复用、可选的 PyPI wheel 辅助服务。它转发 PyPI 当前原始索引 +元数据,保留版本、依赖要求、撤回标记、哈希及额外索引设置。wheel 只有在匹配 +当前原始发布元数据的文件名、大小、SHA256 和 HTTPS 来源,并校验完整下载字节后 +才会提供。跨进程锁保护共享下载;元数据不可用或缓存损坏时拒绝使用。 +源码分发包和原始额外索引不经过此 wheel 缓存。 + +v2 设置 `UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple` 和 +`UV_HTTP_TIMEOUT=600`,允许未缓存的 wheel 在发送任何响应正文前完成完整校验。 +安装时限仍为 **1,800 秒**。原始 APT 源和已校验的 APT 索引、uv/Python 发布包 +下载通道保持不变。没有修改任务镜像、包版本要求、官方测试或原生 agent/验证器 +时限。包缓存和新增依赖准备会影响运行时间,因此这不是对模型行为或下载通道 +进行隔离控制的实验。 + +**六道全新 v2 仅安装检查均在任何模型任务前通过**,包括 SDK 初始化、Linux +进程监管和精确任务修订对应的依赖配置。新增五项配置,模型恢复沿用已有配置。 +它们保留原始 Python、包和索引要求,只在 `/tmp` 执行 `pytest --version`, +不执行任务代码或断言。 +[预检证据](../../configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json) +和[启动登记](../../configs/deepseek-harness-dev6-supplement-20261101/launch-evidence.json) +均先于模型运行。六道正式任务均重复执行并通过相同的依赖配置。CLI 安装耗时 **1,690.4 秒**,仍在未变的 1,800 秒时限内;先前预检通过并不意味着后续下载没有延迟。 + +完整 Harbor 测试套件 **301 项通过**,包括 **16 项 wheel 缓存契约测试**。 +使用全新临时 uv 缓存、真实原版 **uv 0.9.5** 的下载冒烟检查也通过, +**模型调用为零**;[探测记录](../../configs/deepseek-harness-dev6-supplement-20261101/transport-v2/transport-probes.json) +保留两个版本的冒烟检查和一次完成校验的分段下载实验。冻结的下载日志包含 **675 条原始发布元数据事件、270 次已校验 wheel 响应:224 次缓存命中、46 次非缓存响应、53 个不同产物**,来自 **12 个辅助服务**,**记录的下载错误为 0**。每次 wheel 响应均有同一辅助服务此前对应的原始发布元数据。 + +## 保留的模型配置及审计 + +| 属性 | 值 | +| --- | --- | +| 上游 | `deepseek-ai/deepseek-harness` `0.2.2-alpha.2`,源码 `f3803c96974ad4a94195fb97e5c637e9c2a3afe3` | +| 配置 | `sdk-minimal`;与原始运行相同的离线 SDK/runtime 分发包 | +| 模型/API | `deepseek/deepseek-flash`,high 推理,`https://api.deepseek.com/anthropic` | +| 生成设置 | 原生提示词/工具/压缩/请求重试;temperature 和 SDK max_tokens 未设置;原生上限 256,000 | +| 适配器 | 共享冻结源码 `eff073b1601435305155cad92b6ee9e2250fbbdc` | +| 下载通道源码 | `dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3`;已检查登记的文件哈希 | +| 时限 | 每题 agent 900 秒;验证器 900 秒,查询题为 1,800 秒;安装 1,800 秒 | +| 调度 | 原始相对顺序;并发二,每组完成后再推进;每题一个新的正式尝试,无自动整题重试 | +| 验证器重试 | 仅超时重试,最多三次总尝试,输出及原始单次时限不变 | + +冻结的适配器相对原始源码仅修改了 `bootstrap.py` 中经过审核的依赖预检;模型 +驱动、提示词集成、监管器和验证器的字节完全一致。用户的上游检出目录保持不动。 +全新隔离会话不接收之前的输出、诊断、解答或个人配置。其余 **55 道题保持未打开**。 +没有加入 SDK 服务生命周期修复;原始 KV/PyPI 失败保持不变。 + +11 月 1 日检查的官方[agent 集成指南](https://api-docs.deepseek.com/quick_start/agent_integrations/) +仍将该别名映射到 DeepSeek-V4.1-Flash-1017,见 +[别名证据](../../configs/deepseek-harness-dev6-supplement-20261101/model-alias-check.json)。 +它是可变别名,不能证明跨日期使用相同后端检查点。Pi 使用 OpenAI 兼容接口, +输出上限为 384,000 token。其采样器/WAL 成绩存在无效/未确认的历史边界,流水线 +并行通过结果带有参考资料接触限制。在这组选出的六道题上,Harness 和 Pi 各通过 **4 题**;Harness 新通过查询优化、失去流水线并行通过,仍保留 Pi 的参考资料接触限制。混合覆盖中 **32 道共同有效计分题**上,Harness 通过 **25 题**,Pi 通过 **26 题**。新增通过为 DNA 组装和查询优化;失去通过为 KV、PyPI 和流水线并行。这些失去通过的题分别带有 KV/PyPI 生命周期及 Pi 流水线参考资料接触限制。 + +冻结后扫描了 **98 次工具调用**,并查看 **3 个候选及对应输出**:CLI 的两次调用访问公开 MNIST 数据,流水线的一次调用检查根目录、Python 路径及环境。未在这些记录中观察到基准参考测试/解答。所查看的环境没有以敏感凭证名称命名的条目;精确匹配在 **590 个冻结文件中发现该已授权服务商密钥的文件数为 0**。 这种定向审计不能穷尽证明不存在资料接触或数据外传。 +SDK 已知用量为 **1,298,682 个输入 token**,其中 **1,242,112 个缓存 token**,以及 **79,429 个输出 token**。六道试次均报告 `usage_is_partial=false`;费用仍未知。 缓存 token 已包含在输入中,不再重复相加;这些数据是 SDK 已知用量小计,未查询服务商账单。正式队列从 **2026-11-01 15:50:21.232772 UTC** 运行至 **16:55:21.072680 UTC**:墙钟耗时 **3,899.8 秒**,agent 累计耗时 **1,744.3 秒**。队列时间包含安装、验证及组间等待,不包含之前的预检和下载探测。 + +```text +v2 plan SHA256: 69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93 +freeze SHA256: ee6a3c6a4f57b6da2ffcc04e81e54e888949285054aff8e7a218ce0f330671f3 +results SHA256: 412a0d0acd266d5573ff722a2e5376ab0264b457821d0d603cf8b3f0e663d067 +preflight evidence SHA256: e973e6d280171287c8c60b7888834b56259f897aac67e8ffa6c23daad39e3e06 +original results SHA256: e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396 +source archive SHA256: 0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4 +distribution SHA256: d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6 +runtime SHA256: 28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2 +wheel-cache source SHA256: ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74 +wheel audit SHA256: ef8385be4d1387b72014dca1e89b1281fed347714f9bd3e5f9e050b501e7366a +``` + +## 复现 + +使用共享冻结的适配器检出目录和登记的本地产物。编排器调用标准 +`python -m harbor_adapter run --config ...` 入口,不创建每个任务独立的适配器副本。 +发布的配置保留本地绝对路径,复现时应使用等价的固定产物及新任务名。 +不要覆盖或重放已登记的任务。 + +```bash +export UV_PROJECT_ENVIRONMENT="$PWD/jobs/dsh-harbor-venv" +export UV_CACHE_DIR="$PWD/.cache/uv" +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json \ + --phase preflight +uv run --frozen --project benchmarks/harbor \ + python benchmarks/harbor/scripts/run_comparison.py \ + --plan benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json \ + --phase model +``` + +已授权的服务商密钥仅通过进程环境提供。英文和简体中文报告包含相同的结果、 +限制、表格、标识符、命令和哈希。 diff --git a/benchmarks/harbor/results/deepseek-harness-dev6-supplement-20261101.json b/benchmarks/harbor/results/deepseek-harness-dev6-supplement-20261101.json new file mode 100644 index 0000000..17d9ec4 --- /dev/null +++ b/benchmarks/harbor/results/deepseek-harness-dev6-supplement-20261101.json @@ -0,0 +1,2054 @@ +{ + "schema_version": 1, + "experiment": "deepseek-harness-dev6-supplement-20261101-v2", + "audited_at": "2026-11-01T16:57:53.619037+00:00", + "frozen_at": "2026-11-01T16:56:07.773103+00:00", + "freeze_manifest": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/outputs-freeze.json", + "freeze_sha256": "ee6a3c6a4f57b6da2ffcc04e81e54e888949285054aff8e7a218ce0f330671f3", + "registration": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json", + "registration_sha256": "69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93", + "scope": "Outcome-selected supplemental development experiment explicitly requested after auditing the original run. Do not replace original rewards or claim an independent clean benchmark. No prior outputs, solutions, official assertions or diagnostics are passed to the model. Inspect only operational metadata until all supplemental outputs are frozen. The other 55 tasks remain unopened.", + "model_policy": { + "model": "deepseek/deepseek-flash", + "provider": "deepseek-official", + "api": "anthropic-messages", + "base_url": "https://api.deepseek.com/anthropic", + "reasoning_effort": "high", + "profile": "sdk-minimal", + "temperature": "unset", + "sdk_max_tokens": "unset", + "upstream_default_max_tokens": 256000, + "alias": "Official guide checked 2026-11-01 maps the mutable deepseek-flash alias to DeepSeek-V4.1-Flash-1017." + }, + "budget_policy": "Original task-native timeouts/resources; 1800-second setup. Upstream sdk-minimal prompts, tools, compaction, retries, and generation defaults. No injected turn limit, internal time budget, or finalization reminders. Pi used OpenAI-compatible transport with 384000-token model metadata; this is not a matched-generation-budget causal harness comparison.", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "sources": { + "adapter": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "upstream": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "upstream_origin": "https://github.com/deepseek-ai/deepseek-harness.git", + "upstream_version": "0.2.2-alpha.2", + "source_archive_sha256": "0a851c705c243519cc573d077d8cf23295591498825d82b795b43b37255d43e4", + "runtime_sha256": "28e1eaeac3c594d2ce74c67c4f1e179c58ccf3d4dbb8c0faa719bec5745805f2", + "profile": "sdk-minimal", + "transport": "dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3" + }, + "original_experiment": { + "path": "benchmarks/harbor/results/deepseek-harness-dev34-20261030.json", + "sha256": "e94ca1e201d032617e9d4649f2cbfbb60127c1257e7a0f3b2696fea808288396", + "unchanged_summary": { + "registered_jobs": 34, + "model_executions": 32, + "passed": 22, + "failed": 6, + "unscored": 6, + "analytical_classifications": { + "failed": 6, + "passed": 22, + "unscored_verifier_error": 3, + "unscored_setup_error": 2, + "unscored_execution_error": 1 + }, + "native_rewards": { + "one": 22, + "zero": 10, + "null": 2 + }, + "standard_summary_statuses": { + "execution_error": 4, + "scored": 28, + "setup_failed": 2 + }, + "agent_timeouts": 3, + "whole_trial_retries": 0, + "extra_verifier_attempts": 3, + "service_lifecycle_caveat_failures": 2, + "accepted_official_cases": { + "total": 118, + "passed": 107, + "failed": 11, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "unscored_query_official_cases": { + "passed": 2, + "failed": 4 + } + }, + "policy": "Preserve original results. These six fresh attempts were separately authorized after the original audit; do not overwrite or choose a better score." + }, + "summary": { + "registered_jobs": 6, + "model_executions": 6, + "passed": 4, + "failed": 2, + "unscored": 0, + "analytical_classifications": { + "passed": 4, + "failed": 2 + }, + "native_rewards": { + "one": 4, + "zero": 2, + "null": 0 + }, + "standard_summary_statuses": { + "scored": 6 + }, + "agent_timeouts": 0, + "whole_trial_retries": 0, + "extra_verifier_attempts": 0, + "accepted_official_cases": { + "passed": 34, + "failed": 3, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0, + "total": 37 + } + }, + "accounting_policy": "Native rewards, exceptions and standard summaries are unchanged. Confirmed timeout outputs with completed official tests remain scoreable; setup failures, terminal provider transport failure and verifier dependency failures remain analytically null. Service-lifecycle failures are retained as failures of this measured SDK configuration. This is not a complete 34-task pass rate or a causal agent-quality comparison.", + "mixed_coverage": { + "original_accepted_attempts": 28, + "new_attempts": 6, + "passed": 26, + "failed": 8, + "unscored": 0, + "policy": "Descriptive coverage using the 28 original accepted attempts plus exactly the six authorized new attempts. Outcome-selected mixed-date sample, not an independent benchmark or best-of-two score." + }, + "historical_comparison": { + "baseline": { + "path": "benchmarks/harbor/results/pi-deepseek-flash-dev34-20261024.json", + "sha256": "b85d3bf2b0ea9661940f0e6b08e3bea71b2c41ae34da5c89f9ecd5197c6c8ff5" + }, + "supplement_common_accepted_tasks": 6, + "supplement_deepseek_passes": 4, + "supplement_pi_passes_on_common": 4, + "mixed_common_accepted_tasks": 32, + "mixed_deepseek_passes_on_common": 25, + "mixed_pi_passes_on_common": 26, + "mixed_gained_passes": [ + "terminal-bench/dna-assembly", + "terminal-bench/query-optimize" + ], + "mixed_lost_passes": [ + "terminal-bench/kv-store-grpc", + "terminal-bench/pypi-server", + "terminal-bench/torch-pipeline-parallelism" + ], + "caveats": "The KV/PyPI losses have the confirmed SDK service-lifecycle caveat. The pipeline loss compares against a Pi pass exposed to public reference tests/solution. WAL passes here but was unscored in Pi and is not a paired gain; Pi sampler/WAL have invalid/unverified legacy boundaries. Different dates, mutable alias, Messages versus OpenAI-compatible endpoints, 256000 versus 384000 output ceilings, new dependency preparation/cache, and outcome selection prevent causal harness attribution." + }, + "usage": { + "n_input_tokens_known": 1298682, + "n_cache_tokens_known": 1242112, + "n_output_tokens_known": 79429, + "input_includes_cache": true, + "cost_usd": null, + "usage_is_partial": false, + "partial_usage_tasks": [], + "policy": "Known SDK usage subtotals only. Interrupted requests and failed native retries can lack usage. Cache tokens are already included in input; do not add them again. No provider billing or price-derived cost is asserted." + }, + "timing": { + "first_job_started_at": "2026-11-01T15:50:21.232772+00:00", + "last_job_finished_at": "2026-11-01T16:55:21.072680+00:00", + "queue_wall_seconds": 3899.839908, + "summed_agent_seconds": 1744.285722, + "interpretation": "Formal queue wall time includes setup, verification and pair barriers, but excludes earlier installation-only checks and transport probes. This supplement had no verifier retries or terminal API failures. Different dates and preparation prevent a controlled latency comparison." + }, + "validation": { + "audited_at": "2026-11-01T16:56:03.016491+00:00", + "final_plan_validation_passed": true, + "source_task_and_configuration_hashes_match": true, + "all_six_preflights_before_first_model_job": true, + "pair_barriers_passed": true, + "registered_jobs_finished": 6, + "model_executions": 6, + "formal_dependency_profiles_passed": 6, + "active_task_containers_after_run": 0, + "v2_images_verified": 12, + "original_frozen_files_rehashed_unchanged": 2506, + "first_job_started_at": "2026-11-01T15:50:21.232772+00:00", + "last_job_finished_at": "2026-11-01T16:55:21.072680+00:00", + "frozen_files_rehashed": 590, + "process_boundaries_confirmed_before_verification": 6, + "post_agent_end_events": 0, + "events_at_or_after_verifier_start": 0, + "final_harbor_tests_passed": 301, + "pypi_cache_contracts_passed": 16, + "native_uv_0_9_5_transport_smoke_passed": true + }, + "preparation": { + "initial": { + "schema_version": 1, + "audited_at": "2026-11-01T15:26:59.206575+00:00", + "original_plan": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment.json", + "sha256": "cb45a79c8c0b128d6807118a656afc32357de40c250bb6064b9d46798a4232ac" + }, + "summary": { + "installation_passed": 3, + "installation_failed": 1, + "installation_not_started": 2, + "model_executions": 0, + "official_verifier_executions": 0 + }, + "reason": "PyTorch model recovery reached the 1800-second setup limit while downloading original verifier dependencies. After pair 2 completed, stop the controller before pair 3 and register verified PyPI wheel transport. No running trial was interrupted by this boundary stop.", + "boundary_stop": { + "path": "jobs/dsh-supplement-preflight-boundary-stop.json", + "sha256": "1a058982408952541e9a98968d4c7e85f36ba050fe6693b1ef994b08ab3b5800" + }, + "controller_exit_code": 130, + "active_task_containers_after_stop": 0, + "trials": [ + { + "task": "terminal-bench/multi-source-data-merger", + "trial": "multi-source-data-merger__A9ZBWfv", + "status": "setup_passed", + "exception_type": null, + "verifier_profile": { + "profile": "multi-source-data-merger-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T14:37:39.038730Z", + "finished_at": "2026-11-01T14:39:16.479499Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-01-20261101.json", + "sha256": "7907d0ca8916a3f8ac35a763309bf13eef2e06e3a9477f4353d8ef1d2450f79e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/result.json", + "sha256": "c3efdc8bb4fb1048f32aa4c88105b8c1a08262e98c2fbe329d41dd09853ae408" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/multi-source-data-merger__A9ZBWfv/agent/bootstrap.json", + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-cli", + "trial": "pytorch-model-cli__WHD5rDb", + "status": "setup_passed", + "exception_type": null, + "verifier_profile": { + "profile": "pytorch-model-cli-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T14:37:39.001514Z", + "finished_at": "2026-11-01T14:52:46.672756Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-01-20261101.json", + "sha256": "7907d0ca8916a3f8ac35a763309bf13eef2e06e3a9477f4353d8ef1d2450f79e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/result.json", + "sha256": "1061d55f449577a08bfb866a4572bf1cda8efeba9e58e731b5d70a7972df0d22" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-20261101/pytorch-model-cli__WHD5rDb/agent/bootstrap.json", + "sha256": "a3c0057981c5d3db5eefa64571d2bcc3354657ae9602961506ab91df8197e94a" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "trial": "pytorch-model-recovery__LD8MzLU", + "status": "setup_failed", + "exception_type": "AgentSetupTimeoutError", + "verifier_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "incomplete" + }, + "agent_setup": { + "started_at": "2026-11-01T14:53:24.216340Z", + "finished_at": "2026-11-01T15:23:24.570179Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-02-20261101.json", + "sha256": "61fa3336d6839abc032ab09e74b840c8b17301726df6d2955dfef9bb58eccd82" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/result.json", + "sha256": "57b535c90e77e1bfb129bdb32e45f29e825132e0dfde847fcab2be97346db9b3" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/pytorch-model-recovery__LD8MzLU/agent/bootstrap.json", + "sha256": "569634e2774a66f6a021d30250a11667067e2fb7503586e9cffd6f8eadacdb75" + } + ] + }, + { + "task": "terminal-bench/query-optimize", + "trial": "query-optimize__nZVwCmT", + "status": "setup_passed", + "exception_type": null, + "verifier_profile": { + "profile": "query-optimize-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T14:53:24.209263Z", + "finished_at": "2026-11-01T14:54:26.697416Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-orchestration/dsh-sup6-install-pair-02-20261101.json", + "sha256": "61fa3336d6839abc032ab09e74b840c8b17301726df6d2955dfef9bb58eccd82" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/result.json", + "sha256": "bcddf59e8ca8e1042bc40c41d2928ef8a7ec7d419cc2fe234f42ae00ba304de2" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-20261101/query-optimize__nZVwCmT/agent/bootstrap.json", + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826" + } + ] + } + ] + }, + "final": { + "schema_version": 1, + "audited_at": "2026-11-01T15:49:52.094028+00:00", + "experiment": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json", + "sha256": "69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93" + }, + "adapter_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "distribution_sha256": "d8f6157c880dfa5ae65b6c176dd86b1e32f5f31709f8f29fa7170db45d2fb9d6", + "model_jobs_started": 0, + "summary": { + "installation_passed": 6, + "sdk_initialization_passed": 6, + "supervision_probe_passed": 6, + "verifier_dependency_profiles_passed": 6, + "verifier_dependency_profiles_not_configured": 0, + "model_executions": 0 + }, + "trials": [ + { + "task": "terminal-bench/multi-source-data-merger", + "task_ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "job": "dsh-sup6-install-pair-01-cache-20261101", + "trial": "multi-source-data-merger__kDiQNmh", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "multi-source-data-merger-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:30:37.424635Z", + "finished_at": "2026-11-01T15:32:11.448362Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-01-cache-20261101.json", + "sha256": "866efbc2def799a1e374e0e47e4b6f596f2b9c7596525b0c53989afab2656fa0" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/manifest.json", + "sha256": "a1c0a37704126265f30150d7ab8d049adcbef97964b3ca57cc64c927dec3c50e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/summary.json", + "sha256": "7e60fc4d3c5a8ac06c48f630b8359cce05366334fd479d196c49e518f824ed56" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/result.json", + "sha256": "c2457b2afd8de269bfac9c2c67a0c120d1f2604bee58da87f3f01fef3bac4e37" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/multi-source-data-merger__kDiQNmh/agent/bootstrap.json", + "sha256": "aa9fa97bcb943c0408ee2557bce7e13bd0f2782ee383b5ff1df2450b2d7b172d" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-cli", + "task_ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "job": "dsh-sup6-install-pair-01-cache-20261101", + "trial": "pytorch-model-cli__dzNhbDJ", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "pytorch-model-cli-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:30:37.388855Z", + "finished_at": "2026-11-01T15:34:17.660530Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-01-cache-20261101.json", + "sha256": "866efbc2def799a1e374e0e47e4b6f596f2b9c7596525b0c53989afab2656fa0" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/manifest.json", + "sha256": "a1c0a37704126265f30150d7ab8d049adcbef97964b3ca57cc64c927dec3c50e" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101-input/summary.json", + "sha256": "7e60fc4d3c5a8ac06c48f630b8359cce05366334fd479d196c49e518f824ed56" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/result.json", + "sha256": "942b30020eaf1fb96c31f6b9ff272eaf32364f08666f2ff3cb4388e4d089d486" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-01-cache-20261101/pytorch-model-cli__dzNhbDJ/agent/bootstrap.json", + "sha256": "0d3b19888676a0f8c1bd7b850d8a349a49b07f1cebc2e1780e678948be935a60" + } + ] + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "task_ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "job": "dsh-sup6-install-pair-02-cache-20261101", + "trial": "pytorch-model-recovery__XnktXzC", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:34:54.988540Z", + "finished_at": "2026-11-01T15:38:52.082773Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-02-cache-20261101.json", + "sha256": "85ad71a6a8dd6e53fd18764b3eb819e725a00d1fb6f640e130d583aa095a3678" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/manifest.json", + "sha256": "bf7a0c43b592960af76176b8715b8440129886b6becc68f51dce7c2dafb06bcf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/summary.json", + "sha256": "40582b6a4c73c6f9ad84f1e454e3072b5818e5b4213e08082f2b88a3b88f0aa1" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/result.json", + "sha256": "ef17615f7c6a6c91d075791d90ad56adab7e245eb7bda5ce846bbfbf904f2366" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/pytorch-model-recovery__XnktXzC/agent/bootstrap.json", + "sha256": "638922b88c59516844637223aff41c51866cf5e21824c22d7fd7116ebfb1e0fc" + } + ] + }, + { + "task": "terminal-bench/query-optimize", + "task_ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "job": "dsh-sup6-install-pair-02-cache-20261101", + "trial": "query-optimize__mjkeDjx", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "query-optimize-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:34:54.987141Z", + "finished_at": "2026-11-01T15:35:49.586138Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-02-cache-20261101.json", + "sha256": "85ad71a6a8dd6e53fd18764b3eb819e725a00d1fb6f640e130d583aa095a3678" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/manifest.json", + "sha256": "bf7a0c43b592960af76176b8715b8440129886b6becc68f51dce7c2dafb06bcf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101-input/summary.json", + "sha256": "40582b6a4c73c6f9ad84f1e454e3072b5818e5b4213e08082f2b88a3b88f0aa1" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/result.json", + "sha256": "f1e90e69169bf7cfee37a68f7915b981e4cff370caa457d9ba9dffaaa431b7cf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-02-cache-20261101/query-optimize__mjkeDjx/agent/bootstrap.json", + "sha256": "1463562f8a5bf45a5156a7d3fb1283b26fac0a33db2574d396c0f3f18b454826" + } + ] + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "task_ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "job": "dsh-sup6-install-pair-03-cache-20261101", + "trial": "torch-pipeline-parallelism__mXeoYMQ", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "torch-pipeline-parallelism-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:39:41.683509Z", + "finished_at": "2026-11-01T15:48:44.616122Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-03-cache-20261101.json", + "sha256": "2ea8ddedb98188185000466de388c6d4c62c565ebb5a6d457b074ac1adad86b3" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/manifest.json", + "sha256": "a79d0cfc3f8c90c299dfdbf2f245115c51924cb56026116b53ccd54bc44abf9d" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/summary.json", + "sha256": "6ac52365e43f74afd60a3424faf8d05b7df928abeca72c1559362664ecd8b348" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/result.json", + "sha256": "d29a589b7595b286187ff3dea2ec985151dfda8cb19da12fee759f75c59771bf" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-pipeline-parallelism__mXeoYMQ/agent/bootstrap.json", + "sha256": "a9460a42b2e140bbe0bee9d4a2226d95a34d556d4cdf92a6b310b8b3a72c3bca" + } + ] + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "task_ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "job": "dsh-sup6-install-pair-03-cache-20261101", + "trial": "torch-tensor-parallelism__6VSbzJS", + "installation": "passed", + "sdk_initialization": "passed", + "process_supervision_probe": "passed", + "verifier_dependency_profile": { + "profile": "torch-tensor-parallelism-v1", + "status": "passed" + }, + "agent_setup": { + "started_at": "2026-11-01T15:39:41.684726Z", + "finished_at": "2026-11-01T15:48:43.633194Z" + }, + "evidence": [ + { + "path": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-install-pair-03-cache-20261101.json", + "sha256": "2ea8ddedb98188185000466de388c6d4c62c565ebb5a6d457b074ac1adad86b3" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/manifest.json", + "sha256": "a79d0cfc3f8c90c299dfdbf2f245115c51924cb56026116b53ccd54bc44abf9d" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101-input/summary.json", + "sha256": "6ac52365e43f74afd60a3424faf8d05b7df928abeca72c1559362664ecd8b348" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/result.json", + "sha256": "4862f8fa3a7b1038b169aa97a54f0892917ad45bdb769a3fb25fba62ade08979" + }, + { + "path": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-install-pair-03-cache-20261101/torch-tensor-parallelism__6VSbzJS/agent/bootstrap.json", + "sha256": "34af5d820280f7c297e211a7e512cc9101f09ab3908a0b7bf5ded9486eb61085" + } + ] + } + ] + }, + "launch": { + "schema_version": 1, + "authorized_launch_registered_at": "2026-11-01T15:49:52.188939+00:00", + "repository_ref": "0c587c016fc0757f08b223a1976fae9e800d352b", + "adapter_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "transport_ref": "dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3", + "authorization": "The maintainer requested rerunning the six unscored tasks on 2026-11-01. Prior explicit confirmation authorizes use of the existing DeepSeek provider credential without publishing it.", + "phase": "model", + "model_jobs_started_at_registration": 0, + "model_runs_authorized": 6, + "whole_trial_retries": 0, + "preflight": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/preflight-evidence.json", + "sha256": "e973e6d280171287c8c60b7888834b56259f897aac67e8ffa6c23daad39e3e06" + }, + "plan": { + "path": "benchmarks/harbor/configs/deepseek-harness-dev6-supplement-20261101/experiment-v2.json", + "sha256": "69eb8272f3fcd125e062387f7f725c80f12b56b8a622d6a616d23f8e0e174a93" + }, + "credential_handling": "Read the authorized local DeepSeek provider credential into the child environment only. No key is written to this record.", + "output_policy": "Freeze all six outputs before inspecting new model trajectories or hidden assertions." + }, + "transport_amendment": "Use UV_DEFAULT_INDEX=http://127.0.0.1:18744/simple to forward the current original PyPI index, preserving versions, requirements, yanks, hashes and extra-index settings. Serve original wheels only after checking current PyPI release filename/size/SHA256/HTTPS origin and full downloaded bytes. Existing ranged downloader, cross-process file locks, and persistent public provenance log. Set UV_HTTP_TIMEOUT=600 for a cold wheel to finish verification before any body is served. Setup remains 1800 seconds; task-native agent/verifier limits and official scripts are unchanged. Original APT and uv/Python release transports remain identical. No model output or task is replayed automatically.", + "transport_probes": { + "recorded_at": "2026-11-01T15:28:42.170928+00:00", + "range_download": { + "filename": "torch-2.7.1-cp313-cp313-manylinux_2_28_x86_64.whl", + "path": ".cache/nanopy-harbor/pypi-ranges/blobs/sha256/d632f5417b6980f61404a125b999ca6ebd0b8b4bbdbb5fbbba44374ab619a412", + "size": 821033192, + "sha256": "d632f5417b6980f61404a125b999ca6ebd0b8b4bbdbb5fbbba44374ab619a412", + "elapsed_seconds": 357.6746226220057, + "bytes_per_second": 2295475.105226228, + "source": "https://pypi.org/pypi/torch/2.7.1/json", + "purpose": "Host-side verified transport probe; not injected into the active preflight." + }, + "initial_native_uv_smoke": { + "checked_at": "2026-11-01T15:19:10.795930+00:00", + "command": [ + "jobs/uv-0.9.5-probe/uv", + "tool", + "run", + "--python", + "/home/minix/.local/share/uv/python/cpython-3.13.15-linux-x86_64-gnu/bin/python3.13", + "--with", + "pytest==8.4.1", + "--with", + "pytest-json-ctrf==0.3.5", + "pytest", + "--version" + ], + "uv_version": "0.9.5", + "script_sha256": "8369dfbba73fb6410b9dbef79d7ddcd4d1190fef90fbf35d58a91ea41f53fbb1", + "exit_code": 0, + "stdout": "pytest 8.4.1", + "model_calls": 0, + "fresh_uv_cache": true, + "default_index": "Loopback proxy forwarding current original PyPI metadata and verified wheels." + }, + "final_native_uv_smoke": { + "checked_at": "2026-11-01T15:26:24.601919+00:00", + "command": [ + "/mnt/ubuntu/home/omarchy_data/nanoPyCodeAgent/jobs/uv-0.9.5-probe/uv", + "tool", + "run", + "--python", + "/home/minix/.local/share/uv/python/cpython-3.13.15-linux-x86_64-gnu/bin/python3.13", + "--with", + "pytest==8.4.1", + "--with", + "pytest-json-ctrf==0.3.5", + "pytest", + "--version" + ], + "uv_version": "0.9.5", + "script_sha256": "ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74", + "probe_sha256": "02d43b929e97cb8a73c32525203cb9fa439294aa3dbfb55e804a41c009483237", + "exit_code": 0, + "stdout": "pytest 8.4.1", + "model_calls": 0, + "fresh_uv_cache": true, + "audit_log": "jobs/dsh-pypi-uv-smoke-audit-events.jsonl", + "audit_log_sha256": "330958a0b95493071fc429c43ed8c4f25a96ff648b88ff4531dc29fd9976b796", + "default_index": "Loopback proxy forwarding current original PyPI metadata and verified wheels." + }, + "validation": { + "full_harbor_tests_passed": 301, + "pypi_cache_tests_passed": 16 + }, + "source_ref": "dc32b26194a06e7b070c5e8876d26ea0fcf3a7d3", + "source_files_sha256": { + "benchmarks/harbor/scripts/pypi_wheel_cache.py": "ced6bd7c1bf2b6ac1964c39d057708ba0e52c37c61a609651c49105ea145ea74", + "benchmarks/harbor/scripts/cache_image.py": "11ddde3fe86a6e4bfdefe7c3578794df98d237f701e40f219d76a541911c9f0c" + } + } + }, + "wheel_transport_audit": { + "path": ".cache/nanopy-harbor/pypi-ranges/audit/dsh-sup6-cache-20261101.jsonl", + "sha256": "ef8385be4d1387b72014dca1e89b1281fed347714f9bd3e5f9e050b501e7366a", + "events": { + "ready": 12, + "release_metadata": 675, + "wheel": 270 + }, + "verified_wheel_responses": 270, + "cache_hits": 224, + "cold_responses": 46, + "unique_verified_artifacts": 53, + "every_wheel_has_matching_prior_original_release_metadata": true, + "artifacts": [ + { + "project": "pygments", + "version": "2.21.0", + "filename": "pygments-2.21.0-py3-none-any.whl", + "size": 1250147, + "sha256": "2363c69b61c4a97c838da3b130dcd6468f4848992b21a82f2a63ec34377137d9", + "source_url": "https://files.pythonhosted.org/packages/71/46/17f022dd3e953bf20a04a028a21ec746d942f8d2af30fa0f124fa0e6a684/pygments-2.21.0-py3-none-any.whl" + }, + { + "project": "tzdata", + "version": "2026.6", + "filename": "tzdata-2026.6-py2.py3-none-any.whl", + "size": 348210, + "sha256": "0a10eec6e13ef16b627fabed9a5f2ba61a62b1c67f327305f4e7e412a70b4aec", + "source_url": "https://files.pythonhosted.org/packages/0a/66/9a06c2227fedfd4919f0eaaa8b94377d9730882a7e4f95823c9a331ab5e1/tzdata-2026.6-py2.py3-none-any.whl" + }, + { + "project": "pytest", + "version": "8.4.1", + "filename": "pytest-8.4.1-py3-none-any.whl", + "size": 365474, + "sha256": "539c70ba6fcead8e78eebbf1115e8b589e7565830d7d006a8723f19ac8a0afb7", + "source_url": "https://files.pythonhosted.org/packages/29/16/c8a903f4c4dffe7a12843191437d7cd8e32751d5de349d45d3fe69544e87/pytest-8.4.1-py3-none-any.whl" + }, + { + "project": "pytz", + "version": "2026.6", + "filename": "pytz-2026.6-py2.py3-none-any.whl", + "size": 506321, + "sha256": "7fed3fee8fd6bd4a2b7b75a8d56021d0624127bf69bd5691f115db929ac570b3", + "source_url": "https://files.pythonhosted.org/packages/b6/25/564f3fc5b70c1398d706c51f2b1470d0ea0b1b2f96fee67877b6f5bdf3b6/pytz-2026.6-py2.py3-none-any.whl" + }, + { + "project": "pytest-json-ctrf", + "version": "0.3.5", + "filename": "pytest_json_ctrf-0.3.5-py3-none-any.whl", + "size": 6837, + "sha256": "e82fd1d69be2f92385bc33540063e5ad7b17b36de67764c84f3ceb9815a895e9", + "source_url": "https://files.pythonhosted.org/packages/51/a8/d58637d10f70ae14a1ad18cb37f4aa32d5593510213d5f929892cb1c9c6d/pytest_json_ctrf-0.3.5-py3-none-any.whl" + }, + { + "project": "pandas", + "version": "2.3.3", + "filename": "pandas-2.3.3-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", + "size": 12306371, + "sha256": "318d77e0e42a628c04dc56bcef4b40de67918f7041c2b061af1da41dcff670ac", + "source_url": "https://files.pythonhosted.org/packages/15/07/284f757f63f8a8d69ed4472bfd85122bd086e637bf4ed09de572d575a693/pandas-2.3.3-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "packaging", + "version": "26.3", + "filename": "packaging-26.3-py3-none-any.whl", + "size": 129956, + "sha256": "d7193f7c8e4e93f444fde0262bf90af30e16fa0ad0ad44cb553c87339b23cd1c", + "source_url": "https://files.pythonhosted.org/packages/63/34/ba1c580383c9eada3711951fef0795c80b829a078d72188184bcab9dd527/packaging-26.3-py3-none-any.whl" + }, + { + "project": "pluggy", + "version": "1.7.0", + "filename": "pluggy-1.7.0-py3-none-any.whl", + "size": 27082, + "sha256": "7dd7b0d8832ba3cb632c306926ded123429211b83641b35dc5c41ad2d34f9bec", + "source_url": "https://files.pythonhosted.org/packages/40/9e/2b38731e0fc536806f16490e1a12d7f0dc2a1235aa8cc07bcc75416a7daa/pluggy-1.7.0-py3-none-any.whl" + }, + { + "project": "numpy", + "version": "2.5.4", + "filename": "numpy-2.5.4-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", + "size": 16718044, + "sha256": "6ffa07666f8da0eef81d149934a626d0d95fbd6838432a33e66245423a9062c0", + "source_url": "https://files.pythonhosted.org/packages/da/24/3bd070f3269dc609d8f26b2643f62ef91bb415841c0b294805aaf7fe06da/numpy-2.5.4-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "iniconfig", + "version": "2.3.1", + "filename": "iniconfig-2.3.1-py3-none-any.whl", + "size": 7552, + "sha256": "9121e2c1fdb355232495be3194c8dfe87ccc2d5dee45947b78e68f499790d7a7", + "source_url": "https://files.pythonhosted.org/packages/56/43/4ca9e49d27a1fcf6bece6f6aec0ea46bb9112489b93d4b688fb415457bdb/iniconfig-2.3.1-py3-none-any.whl" + }, + { + "project": "python-dateutil", + "version": "4.0.0", + "filename": "python_dateutil-4.0.0-py3-none-any.whl", + "size": 92853, + "sha256": "a5e246f79eb80475e691bf93862cead98decd8a498cf53ba0516fcdc0f50dc1c", + "source_url": "https://files.pythonhosted.org/packages/2f/a9/52b89563d154433d79b51ed1eb4842bf7334cb2696891a7fe135618853d7/python_dateutil-4.0.0-py3-none-any.whl" + }, + { + "project": "pyarrow", + "version": "22.0.0", + "filename": "pyarrow-22.0.0-cp313-cp313-manylinux_2_28_x86_64.whl", + "size": 47724601, + "sha256": "e0a15757fccb38c410947df156f9749ae4a3c89b2393741a50521f39a8cf202a", + "source_url": "https://files.pythonhosted.org/packages/1b/8b/5362443737a5307a7b67c1017c42cd104213189b4970bf607e05faf9c525/pyarrow-22.0.0-cp313-cp313-manylinux_2_28_x86_64.whl" + }, + { + "project": "filelock", + "version": "4.2.6", + "filename": "filelock-4.2.6-py3-none-any.whl", + "size": 164378, + "sha256": "fd7d4fca497bf3e3d0945a13261f8e684f2aef12ce7cf76aef26e13bf7388b88", + "source_url": "https://files.pythonhosted.org/packages/b7/8d/3f849fcb79359259737bc2d8ddcaee6149ffb773239a260c9ff20179736f/filelock-4.2.6-py3-none-any.whl" + }, + { + "project": "setuptools", + "version": "84.2.1", + "filename": "setuptools-84.2.1-py3-none-any.whl", + "size": 876263, + "sha256": "ca140b45d5325a9478f880e351484f62fef15f336462e449ca0c03c873d589c2", + "source_url": "https://files.pythonhosted.org/packages/ef/55/6dc6d6bb953bdfc52519a0059592c52b6e3b8ad51cf8256dffceb55b0610/setuptools-84.2.1-py3-none-any.whl" + }, + { + "project": "markupsafe", + "version": "3.0.4", + "filename": "markupsafe-3.0.4-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", + "size": 23017, + "sha256": "434139499bb20b502ed3baa1f169e618f924a97e7a777fea1a49446d80106cf6", + "source_url": "https://files.pythonhosted.org/packages/63/e0/cec6865dfe88cb48fedd4b20aed6af5158e41092adcbf3e028bcc6ec2108/markupsafe-3.0.4-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "opencv-python", + "version": "4.11.0.86", + "filename": "opencv_python-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", + "size": 62986597, + "sha256": "6b02611523803495003bd87362db3e1d2a0454a6a63025dc6658a9830570aa0d", + "source_url": "https://files.pythonhosted.org/packages/2c/8b/90eb44a40476fa0e71e05a0283947cfd74a5d36121a11d926ad6f3193cc4/opencv_python-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl" + }, + { + "project": "torch", + "version": "2.7.1", + "filename": "torch-2.7.1-cp313-cp313-manylinux_2_28_x86_64.whl", + "size": 821033192, + "sha256": "d632f5417b6980f61404a125b999ca6ebd0b8b4bbdbb5fbbba44374ab619a412", + "source_url": "https://files.pythonhosted.org/packages/3a/24/efe2f520d75274fc06b695c616415a1e8a1021d87a13c68ff9dce733d088/torch-2.7.1-cp313-cp313-manylinux_2_28_x86_64.whl" + }, + { + "project": "nvidia-cufft-cu12", + "version": "11.3.0.4", + "filename": "nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 200221632, + "sha256": "ccba62eb9cef5559abd5e0d54ceed2d9934030f51163df018532142a8ec533e5", + "source_url": "https://files.pythonhosted.org/packages/8f/16/73727675941ab8e6ffd86ca3a4b7b47065edcca7a997920b831f8147c99d/nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "nvidia-cudnn-cu12", + "version": "9.5.1.17", + "filename": "nvidia_cudnn_cu12-9.5.1.17-py3-none-manylinux_2_28_x86_64.whl", + "size": 570988386, + "sha256": "30ac3869f6db17d170e0e556dd6cc5eee02647abc31ca856634d5a40f82c15b2", + "source_url": "https://files.pythonhosted.org/packages/2a/78/4535c9c7f859a64781e43c969a3a7e84c54634e319a996d43ef32ce46f83/nvidia_cudnn_cu12-9.5.1.17-py3-none-manylinux_2_28_x86_64.whl" + }, + { + "project": "nvidia-cublas-cu12", + "version": "12.6.4.1", + "filename": "nvidia_cublas_cu12-12.6.4.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 393138322, + "sha256": "08ed2686e9875d01b58e3cb379c6896df8e76c75e0d4a7f7dace3d7b6d9ef8eb", + "source_url": "https://files.pythonhosted.org/packages/af/eb/ff4b8c503fa1f1796679dce648854d58751982426e4e4b37d6fce49d259c/nvidia_cublas_cu12-12.6.4.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "triton", + "version": "3.3.1", + "filename": "triton-3.3.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", + "size": 155673035, + "sha256": "b89d846b5a4198317fec27a5d3a609ea96b6d557ff44b56c23176546023c4240", + "source_url": "https://files.pythonhosted.org/packages/74/1f/dfb531f90a2d367d914adfee771babbd3f1a5b26c3f5fbc458dee21daa78/triton-3.3.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "nvidia-cusparse-cu12", + "version": "12.5.4.2", + "filename": "nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 216561367, + "sha256": "7556d9eca156e18184b94947ade0fba5bb47d69cec46bf8660fd2c71a4b48b73", + "source_url": "https://files.pythonhosted.org/packages/06/1e/b8b7c2f4099a37b96af5c9bb158632ea9e5d9d27d7391d7eb8fc45236674/nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "nvidia-nccl-cu12", + "version": "2.26.2", + "filename": "nvidia_nccl_cu12-2.26.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 201319755, + "sha256": "694cf3879a206553cc9d7dbda76b13efaf610fdb70a50cba303de1b0d1530ac6", + "source_url": "https://files.pythonhosted.org/packages/67/ca/f42388aed0fddd64ade7493dbba36e1f534d4e6fdbdd355c6a90030ae028/nvidia_nccl_cu12-2.26.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "nvidia-nvjitlink-cu12", + "version": "12.6.85", + "filename": "nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", + "size": 19744971, + "sha256": "eedc36df9e88b682efe4309aa16b5b4e78c2407eac59e8c10a6a47535164369a", + "source_url": "https://files.pythonhosted.org/packages/9d/d7/c5383e47c7e9bf1c99d5bd2a8c935af2b6d705ad831a7ec5c97db4d82f4f/nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl" + }, + { + "project": "nvidia-cuda-cupti-cu12", + "version": "12.6.80", + "filename": "nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 8917980, + "sha256": "6768bad6cab4f19e8292125e5f1ac8aa7d1718704012a0e3272a6f61c4bce132", + "source_url": "https://files.pythonhosted.org/packages/49/60/7b6497946d74bcf1de852a21824d63baad12cd417db4195fc1bfe59db953/nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "networkx", + "version": "3.7", + "filename": "networkx-3.7-py3-none-any.whl", + "size": 2142205, + "sha256": "e3fd2c13a7814cee3746340d8d7f8598a67f16a58bf47fb7f8793fab6efca1b0", + "source_url": "https://files.pythonhosted.org/packages/7e/cd/fe58041e9011f307c490e3e17dd48cc516448f7c698a3f2d9d9d65d7e6a8/networkx-3.7-py3-none-any.whl" + }, + { + "project": "typing-extensions", + "version": "4.16.0", + "filename": "typing_extensions-4.16.0-py3-none-any.whl", + "size": 45571, + "sha256": "481caa481374e813c1b176ada14e97f1f67a4539ce9cfeb3f350d78d6370c2e8", + "source_url": "https://files.pythonhosted.org/packages/49/d3/b8441a820a491ddfc024b0b0cf0393375b75ea13866d9c66727e54c2fc80/typing_extensions-4.16.0-py3-none-any.whl" + }, + { + "project": "nvidia-cuda-nvrtc-cu12", + "version": "12.6.77", + "filename": "nvidia_cuda_nvrtc_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl", + "size": 23650380, + "sha256": "35b0cc6ee3a9636d5409133e79273ce1f3fd087abb0532d2d2e8fff1fe9efc53", + "source_url": "https://files.pythonhosted.org/packages/75/2e/46030320b5a80661e88039f59060d1790298b4718944a65a7f2aeda3d9e9/nvidia_cuda_nvrtc_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl" + }, + { + "project": "sympy", + "version": "1.14.0", + "filename": "sympy-1.14.0-py3-none-any.whl", + "size": 6299353, + "sha256": "e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5", + "source_url": "https://files.pythonhosted.org/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl" + }, + { + "project": "nvidia-curand-cu12", + "version": "10.3.7.77", + "filename": "nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 56279010, + "sha256": "a42cd1344297f70b9e39a1e4f467a4e1c10f1da54ff7a85c12197f6c652c8bdf", + "source_url": "https://files.pythonhosted.org/packages/73/1b/44a01c4e70933637c93e6e1a8063d1e998b50213a6b65ac5a9169c47e98e/nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "fsspec", + "version": "2026.9.0", + "filename": "fsspec-2026.9.0-py3-none-any.whl", + "size": 221738, + "sha256": "8dd6e646e99ea382bd85f97a45e6b526a442d79423a7dc673f1e2756d05fcb5f", + "source_url": "https://files.pythonhosted.org/packages/6c/c0/a98505f18594f1bce828bb159cec0fcf9860562f1a2c85913409fc8f3d9e/fsspec-2026.9.0-py3-none-any.whl" + }, + { + "project": "nvidia-nvtx-cu12", + "version": "12.6.77", + "filename": "nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 89276, + "sha256": "b90bed3df379fa79afbd21be8e04a0314336b8ae16768b58f2d34cb1d04cd7d2", + "source_url": "https://files.pythonhosted.org/packages/56/9a/fff8376f8e3d084cd1530e1ef7b879bb7d6d265620c95c1b322725c694f4/nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "nvidia-cusparselt-cu12", + "version": "0.6.3", + "filename": "nvidia_cusparselt_cu12-0.6.3-py3-none-manylinux2014_x86_64.whl", + "size": 156785796, + "sha256": "e5c8a26c36445dd2e6812f1177978a24e2d37cacce7e090f297a688d1ec44f46", + "source_url": "https://files.pythonhosted.org/packages/3b/9a/72ef35b399b0e183bc2e8f6f558036922d453c4d8237dab26c666a04244b/nvidia_cusparselt_cu12-0.6.3-py3-none-manylinux2014_x86_64.whl" + }, + { + "project": "mpmath", + "version": "1.3.0", + "filename": "mpmath-1.3.0-py3-none-any.whl", + "size": 536198, + "sha256": "a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", + "source_url": "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl" + }, + { + "project": "nvidia-cufile-cu12", + "version": "1.11.1.6", + "filename": "nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 1142103, + "sha256": "cc23469d1c7e52ce6c1d55253273d32c565dd22068647f3aa59b3c6b005bf159", + "source_url": "https://files.pythonhosted.org/packages/b2/66/cc9876340ac68ae71b15c743ddb13f8b30d5244af344ec8322b449e35426/nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "jinja2", + "version": "3.1.6", + "filename": "jinja2-3.1.6-py3-none-any.whl", + "size": 134899, + "sha256": "85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67", + "source_url": "https://files.pythonhosted.org/packages/62/a1/3d680cbfd5f4b8f15abc1d571870c5fc3e594bb582bc3b64ea099db13e56/jinja2-3.1.6-py3-none-any.whl" + }, + { + "project": "nvidia-cusolver-cu12", + "version": "11.7.1.2", + "filename": "nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 158229790, + "sha256": "e9e49843a7707e42022babb9bcfa33c29857a93b88020c4e4434656a655b698c", + "source_url": "https://files.pythonhosted.org/packages/f0/6e/c2cf12c9ff8b872e92b4a5740701e51ff17689c4d726fca91875b07f655d/nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "nvidia-cuda-runtime-cu12", + "version": "12.6.77", + "filename": "nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 897690, + "sha256": "ba3b56a4f896141e25e19ab287cd71e52a6a0f4b29d0d31609f60e3b4d5219b7", + "source_url": "https://files.pythonhosted.org/packages/e1/23/e717c5ac26d26cf39a27fbc076240fad2e3b817e5889d671b67f4f9f49c5/nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "idna", + "version": "3.20", + "filename": "idna-3.20-py3-none-any.whl", + "size": 69583, + "sha256": "ab7ae7122974553370f0bdb919e1a960b2cd1bc1ef0276416d896db81c14582c", + "source_url": "https://files.pythonhosted.org/packages/58/a2/bb081bab032533a855d44de1d56f8e8426114ff1ba5d1f07a438a0a654f8/idna-3.20-py3-none-any.whl" + }, + { + "project": "transformers", + "version": "4.55.0", + "filename": "transformers-4.55.0-py3-none-any.whl", + "size": 11267905, + "sha256": "29d9b8800e32a4a831bb16efb5f762f6a9742fef9fce5d693ed018d19b106490", + "source_url": "https://files.pythonhosted.org/packages/1c/93/bcb22fb52ed65084c0199270832aa4cdd4b41296d896f3e7ade188bccb68/transformers-4.55.0-py3-none-any.whl" + }, + { + "project": "safetensors", + "version": "0.8.0", + "filename": "safetensors-0.8.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", + "size": 516040, + "sha256": "fd6f3f93c9a0a7cc2788ee63fb763353d4bd2e89b0751bc78fcf7dda00bea774", + "source_url": "https://files.pythonhosted.org/packages/28/50/f203ff3a3ddfe19308efc83c5a3a29ed02bf786732ec35e68bf9162f3365/safetensors-0.8.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl" + }, + { + "project": "requests", + "version": "2.34.2", + "filename": "requests-2.34.2-py3-none-any.whl", + "size": 73075, + "sha256": "2a0d60c172f83ac6ab31e4554906c0f3b3588d37b5cb939b1c061f4907e278e0", + "source_url": "https://files.pythonhosted.org/packages/a0/f4/c67b0b3f1b9245e8d266f0f112c500d50e5b4e83cb6f3b71b6528104182a/requests-2.34.2-py3-none-any.whl" + }, + { + "project": "certifi", + "version": "2026.10.23", + "filename": "certifi-2026.10.23-py3-none-any.whl", + "size": 135082, + "sha256": "e9b54de76e530bb6cf0ccd036787103acb58fd42ba930ea8bf948ddbaa2f61d1", + "source_url": "https://files.pythonhosted.org/packages/32/d1/a3652fe2c19134d0d15922177821d863cbd7e842d53a21bbe7e8e3a1b5ac/certifi-2026.10.23-py3-none-any.whl" + }, + { + "project": "hf-xet", + "version": "1.8.0", + "filename": "hf_xet-1.8.0-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", + "size": 4292483, + "sha256": "29dab7c904cd1914a0439c6e1306c12c7497c8bd32290fbe563ebf51ca4b6d6d", + "source_url": "https://files.pythonhosted.org/packages/f8/1e/1e8ecc05ef034e9a95c18ed32d2ec07905a7cd2ca611a9d88e25ee1f4850/hf_xet-1.8.0-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl" + }, + { + "project": "huggingface-hub", + "version": "0.36.2", + "filename": "huggingface_hub-0.36.2-py3-none-any.whl", + "size": 566395, + "sha256": "48f0c8eac16145dfce371e9d2d7772854a4f591bcb56c9cf548accf531d54270", + "source_url": "https://files.pythonhosted.org/packages/a8/af/48ac8483240de756d2438c380746e7130d1c6f75802ef22f3c6d49982787/huggingface_hub-0.36.2-py3-none-any.whl" + }, + { + "project": "tokenizers", + "version": "0.21.4", + "filename": "tokenizers-0.21.4-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", + "size": 3115426, + "sha256": "51b7eabb104f46c1c50b486520555715457ae833d5aee9ff6ae853d1130506ff", + "source_url": "https://files.pythonhosted.org/packages/f2/90/273b6c7ec78af547694eddeea9e05de771278bd20476525ab930cecaf7d8/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl" + }, + { + "project": "tqdm", + "version": "4.70.1", + "filename": "tqdm-4.70.1-py3-none-any.whl", + "size": 80199, + "sha256": "c293e525e6fef9c20e8728fd4612df02a0aa31bb5fe91ecd93e123b1b7bffa73", + "source_url": "https://files.pythonhosted.org/packages/a7/03/921a3d3c75785aca9ebfbfcabfbc3a1be12e2ab5265deb026d55a5a3f83e/tqdm-4.70.1-py3-none-any.whl" + }, + { + "project": "pyyaml", + "version": "6.0.3", + "filename": "pyyaml-6.0.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", + "size": 801626, + "sha256": "0f29edc409a6392443abf94b9cf89ce99889a1dd5376d94316ae5145dfedd5d6", + "source_url": "https://files.pythonhosted.org/packages/74/27/e5b8f34d02d9995b80abcef563ea1f8b56d20134d8f4e5e81733b1feceb2/pyyaml-6.0.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "charset-normalizer", + "version": "3.5.2", + "filename": "charset_normalizer-3.5.2-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", + "size": 254660, + "sha256": "7218e8f32b0956cfcd048fd42d9d5779809745ca1d86113ca56f66e7ae1549c4", + "source_url": "https://files.pythonhosted.org/packages/4b/85/41d27f234b82e47c167a5f6c0f62501dc0c640585ff4aba79e08a390336a/charset_normalizer-3.5.2-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "regex", + "version": "2026.9.29", + "filename": "regex-2026.9.29-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", + "size": 807902, + "sha256": "d49c18f1ea294cf4adde2e5ac256e98c82ea9d708462ce4bf799dffa7cfe8a2c", + "source_url": "https://files.pythonhosted.org/packages/9a/8e/c1ed81f55f992f6aa0b699a592a50c1ce9e6d44ff1aee2c14c0537dcef9c/regex-2026.9.29-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "urllib3", + "version": "2.8.0", + "filename": "urllib3-2.8.0-py3-none-any.whl", + "size": 135717, + "sha256": "0cf3cae568d36aa9576b28dfb35f11328f1cb974ca7647d9475ebb86c75ac6e3", + "source_url": "https://files.pythonhosted.org/packages/92/9d/c4e665119135114480843e7ab388fa94d8480650450e6f8e26b70d323a4c/urllib3-2.8.0-py3-none-any.whl" + }, + { + "project": "triton", + "version": "3.3.0", + "filename": "triton-3.3.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", + "size": 156516468, + "sha256": "47bc87ad66fa4ef17968299acacecaab71ce40a238890acc6ad197c3abe2b8f1", + "source_url": "https://files.pythonhosted.org/packages/7d/74/4bf2702b65e93accaa20397b74da46fb7a0356452c1bb94dbabaf0582930/triton-3.3.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl" + }, + { + "project": "torch", + "version": "2.7.0", + "filename": "torch-2.7.0-cp313-cp313-manylinux_2_28_x86_64.whl", + "size": 865046389, + "sha256": "9b52347118116cf3dff2ab5a3c3dd97c719eb924ac658ca2a7335652076df708", + "source_url": "https://files.pythonhosted.org/packages/4b/27/285a8cf12bd7cd71f9f211a968516b07dcffed3ef0be585c6e823675ab91/torch-2.7.0-cp313-cp313-manylinux_2_28_x86_64.whl" + } + ], + "scope": "The optional default-PyPI sidecars across the six final preflights and six formal jobs. Original extra indexes and source distributions bypass this wheel cache. Recorded failures, if any, remain in the event counts." + }, + "exposure_audit": { + "method": "After freezing outputs, targeted scan of all recorded tool-call arguments for benchmark repositories, hidden paths, environment access, and explicit URLs, with paired candidate outputs reviewed. This is not an exhaustive proof of absence of exposure or exfiltration.", + "tool_calls_scanned": 98, + "candidates_reviewed": 3, + "external_benchmark_material_observed": false, + "candidates": [ + { + "task": "terminal-bench/pytorch-model-cli", + "call_id": "call_00_tU9eFlBzWHEuPz74xIFQ0867", + "review": "Downloaded the public MNIST dataset for task self-testing, not Terminal-Bench reference tests or a solution." + }, + { + "task": "terminal-bench/pytorch-model-cli", + "call_id": "call_00_1C7DXlnVOpxyCvqVFgNI9177", + "review": "Downloaded the public MNIST dataset for task self-testing, not Terminal-Bench reference tests or a solution." + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "call_id": "call_00_azBMWCwtE55WZURQIVO71536", + "review": "General root/Python/environment inspection. No hidden-test/reference marker in the paired output; authorized provider credential absent in exact scan of all frozen files." + } + ], + "candidate_record": { + "path": "jobs/dsh-supplement-exposure-candidates.json", + "sha256": "4379aa5edd80aee1a403c52524aca0e49724c7d37372e3b3937d41fddf45609b" + }, + "credential_scan": { + "audited_at": "2026-11-01T16:56:09.912130+00:00", + "method": "Exact byte matching of the previously authorized provider credential; value never printed or persisted by this audit.", + "frozen_files_scanned": 590, + "matching_files": [], + "matching_file_count": 0 + } + }, + "trials": [ + { + "task": "terminal-bench/multi-source-data-merger", + "task_ref": "sha256:70367c38732e1beda7b229968a48d60242277c2fa4db91339c3f064c4c230d49", + "job_name": "dsh-sup6-multi-source-data-merger-cache-20261101", + "trial": "multi-source-data-merger__hV2zT6o", + "arm": "deepseek", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "native_root": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o", + "record": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-multi-source-data-merger-cache-20261101.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 3, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793548442.0703769, + "stop": 1793548443.1704707 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 43.93666313999711 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-11-01T15:50:24.907479Z", + "finished_at": "2026-11-01T15:50:27.202661Z" + }, + "agent_setup": { + "started_at": "2026-11-01T15:50:27.202702Z", + "finished_at": "2026-11-01T15:52:49.399944Z" + }, + "agent_execution": { + "started_at": "2026-11-01T15:52:49.400104Z", + "finished_at": "2026-11-01T15:53:18.976790Z" + }, + "verifier": { + "started_at": "2026-11-01T15:53:19.203288Z", + "finished_at": "2026-11-01T15:54:03.373879Z" + } + }, + "agent_seconds": 29.576686, + "usage": { + "n_input_tokens": 7504, + "n_output_tokens": 3107, + "n_cache_tokens": 6400, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 3, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 3, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 3, + "assistant/message": 3, + "tool/call": 2, + "tool/result": 2, + "step/end": 3, + "turn/end": 1 + }, + "event_count": 26, + "tool_calls": 2, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "exposure_candidates": [], + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 34.336406 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-multi-source-data-merger-cache-20261101.json": "19beb040db143599f69d5bec8e7d8603d904967a39dfd61af052a774ca3b82e1", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/result.json": "7964222816878953a237651d62e8035a0031f3d82e720fb8bd95e8bf0e266808", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/manifest.json": "45923d5f9fcb2414b4403b8176faaf171ea48aa38df654a35585db9d1987636b", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101-input/config.json": "57d3f28a413f43f035a1308429d63405d5108808633e3aa0aa3411532e98e968", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/agent/dsh-events.jsonl": "560ffcbee9aa1d0d4a79d995bb14c03b86ec3c053c01ab5dff68bd460bb8f372", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/test-stdout.txt": "afd6eb483b54b284bd88c964f1cc2a21639044f192cd314df11374ea011a24a7", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-multi-source-data-merger-cache-20261101/multi-source-data-merger__hV2zT6o/verifier/retry-summary.json": "227986f8e26eeea0b1299ea126ca86aab0d0d7e9757aaa4b1ac76c6092bb1c20" + }, + "original_attempt": { + "job_name": "dsh-dev34-multi-source-data-merger-20261030", + "trial": "multi-source-data-merger__4guMJdj", + "native_reward": 0.0, + "native_exception_type": null, + "analytical_classification": "unscored_verifier_error", + "analytical_score": null, + "diagnosis": "Official verifier failed downloading/extracting pyarrow==22.0.0 with UV_HTTP_TIMEOUT=30 before pytest began. Native reward zero retained; analytical score null." + }, + "diagnosis": "Completed normally; all three official output-file, merged-value, and conflict-report checks passed.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "public_task_dataset_download": false, + "environment_inspection": false + }, + "formal_dependency_profile": { + "profile": "multi-source-data-merger-v1", + "status": "passed" + } + }, + { + "task": "terminal-bench/pytorch-model-cli", + "task_ref": "sha256:6f11544dcc81a380dc4611902961f5520259c462d9fceb413419d66ab9a96c79", + "job_name": "dsh-sup6-pytorch-model-cli-cache-20261101", + "trial": "pytorch-model-cli__vEstNM5", + "arm": "deepseek", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "native_root": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5", + "record": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-cli-cache-20261101.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 5, + "failed": 1, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 5, + "failed": 1, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793550588.6888318, + "stop": 1793550622.5733538 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 78.03403291099676 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-11-01T15:50:24.167677Z", + "finished_at": "2026-11-01T15:50:26.455779Z" + }, + "agent_setup": { + "started_at": "2026-11-01T15:50:26.455840Z", + "finished_at": "2026-11-01T16:18:36.809693Z" + }, + "agent_execution": { + "started_at": "2026-11-01T16:18:36.809805Z", + "finished_at": "2026-11-01T16:29:05.209666Z" + }, + "verifier": { + "started_at": "2026-11-01T16:29:05.458518Z", + "finished_at": "2026-11-01T16:30:23.648649Z" + } + }, + "agent_seconds": 628.399861, + "usage": { + "n_input_tokens": 463542, + "n_output_tokens": 18192, + "n_cache_tokens": 448000, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 34, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 34, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 34, + "assistant/message": 34, + "tool/call": 33, + "tool/result": 33, + "step/end": 34, + "turn/end": 1 + }, + "event_count": 212, + "tool_calls": 33, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "exposure_candidates": [], + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 660.719365 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-cli-cache-20261101.json": "60351fe47c843efb3c964a39b2c46e1fe247ef0feaa47f98a7a17c1b1d38d354", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/result.json": "52fb13759ec498d5fa96250033110210e7efcb9295d30475e6b37236059f996c", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/manifest.json": "45923d5f9fcb2414b4403b8176faaf171ea48aa38df654a35585db9d1987636b", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101-input/config.json": "f8f2742c2ddab10e6d9ca235789ce2eb951428e14f44c1b44d722af3745da6a5", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/agent/dsh-events.jsonl": "3f97af2928fc3eb8bb6cc8a053134dd123134d2ff7bc431b6fbdaa1582420f76", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/test-stdout.txt": "1776965440cf6137f1918f1aa073281347d52da046e477ced39824d828970356", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-cli-cache-20261101/pytorch-model-cli__vEstNM5/verifier/retry-summary.json": "0a51e3b4774c831a3230ee69cbe7232f3300f7fd5c7794dc8e955a92ff0a9834" + }, + "original_attempt": { + "job_name": "dsh-dev34-pytorch-model-cli-20261030", + "trial": "pytorch-model-cli__z4gK8xd", + "native_reward": null, + "native_exception_type": "NetworkConnectionError", + "analytical_classification": "unscored_setup_error", + "analytical_score": null, + "diagnosis": "Original uv 0.9.11 installer fetch failed before model execution: curl exit 35, SSL_ERROR_SYSCALL connecting to releases.astral.sh:443. Native NetworkConnectionError retained." + }, + "diagnosis": "Completed normally. Five official checks passed; test_cli_tool_output failed because 14 of 50 tested MNIST predictions differed from the fixed expected values (image 0: 7, expected 2). Artifacts and executable checks passed. This is a scored assertion failure, not a dependency or verifier error.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "public_task_dataset_download": true, + "environment_inspection": false + }, + "formal_dependency_profile": { + "profile": "pytorch-model-cli-v1", + "status": "passed" + } + }, + { + "task": "terminal-bench/pytorch-model-recovery", + "task_ref": "sha256:2e628841cff93290919172398e573e794f34c95d9382b7425adde4364022decc", + "job_name": "dsh-sup6-pytorch-model-recovery-cache-20261101", + "trial": "pytorch-model-recovery__bFj4ZnZ", + "arm": "deepseek", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "native_root": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ", + "record": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-recovery-cache-20261101.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 5, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 5, + "passed": 5, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793550930.5976655, + "stop": 1793550935.3707228 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 61.99433842399594 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-11-01T16:30:58.603526Z", + "finished_at": "2026-11-01T16:31:00.997028Z" + }, + "agent_setup": { + "started_at": "2026-11-01T16:31:00.997066Z", + "finished_at": "2026-11-01T16:32:43.532192Z" + }, + "agent_execution": { + "started_at": "2026-11-01T16:32:43.532313Z", + "finished_at": "2026-11-01T16:34:33.611745Z" + }, + "verifier": { + "started_at": "2026-11-01T16:34:33.870507Z", + "finished_at": "2026-11-01T16:35:36.116697Z" + } + }, + "agent_seconds": 110.079432, + "usage": { + "n_input_tokens": 218370, + "n_output_tokens": 10975, + "n_cache_tokens": 203520, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 14, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 14, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 14, + "assistant/message": 14, + "tool/call": 13, + "tool/result": 13, + "step/end": 14, + "turn/end": 1 + }, + "event_count": 92, + "tool_calls": 13, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "exposure_candidates": [], + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 178.33929 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-pytorch-model-recovery-cache-20261101.json": "be0a7e21e0388145ea0d1dc8efcb9f44a30c91c4f98f5f6edee80893626afbc1", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/result.json": "1bd308679ac767e99b6814dfdc510d99ff51a9797a156ee8989bd8e94210075b", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/manifest.json": "4a239d7d2c769565b765d534fae50638d3bc49476237c72b4b6aa2ccbdacf20d", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101-input/config.json": "8cb93987ef5ecafed86ccf244582bf64dc7f8c62f121b0007482c7aeababd8f0", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/agent/dsh-events.jsonl": "4db152a851b1538a1c1dc0dd28d6a89bce1131fad36cac4aa555b7f297d549b3", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/test-stdout.txt": "36929bebecc3c4ea17c0557f1a45ad5353768b62312508f366c1097fb91cd0d1", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-pytorch-model-recovery-cache-20261101/pytorch-model-recovery__bFj4ZnZ/verifier/retry-summary.json": "5f1d52539b277524eaaa8ed6879ef319d00882f26f696a1fc8a86f23bfa1811e" + }, + "original_attempt": { + "job_name": "dsh-dev34-pytorch-model-recovery-20261030", + "trial": "pytorch-model-recovery__VjqabNQ", + "native_reward": null, + "native_exception_type": "AgentSetupTimeoutError", + "analytical_classification": "unscored_setup_error", + "analytical_score": null, + "diagnosis": "The formal job exceeded the 1800-second setup limit during verifier_dependencies. SDK initialization and supervision probe passed, but the registered dependency profile did not finish. No model execution occurred. The earlier installation-only profile check had passed; native AgentSetupTimeoutError retained." + }, + "diagnosis": "Completed normally; all five official checks passed, including unchanged weights, loading/state-dict equality, and model loss.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "public_task_dataset_download": false, + "environment_inspection": false + }, + "formal_dependency_profile": { + "profile": "pytorch-model-recovery-v1", + "status": "passed" + } + }, + { + "task": "terminal-bench/query-optimize", + "task_ref": "sha256:169496ea6843cb403b0860132675baf7aa0df0ac8b86221068e27d86395260f4", + "job_name": "dsh-sup6-query-optimize-cache-20261101", + "trial": "query-optimize__AEYyxag", + "arm": "deepseek", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "native_root": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag", + "record": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-query-optimize-cache-20261101.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 6, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 6, + "passed": 6, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793551007.5242171, + "stop": 1793551142.013912 + }, + "verifier_attempts": { + "timeout_sec": 1800.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 159.69752336999954 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-11-01T16:30:58.609668Z", + "finished_at": "2026-11-01T16:31:00.993740Z" + }, + "agent_setup": { + "started_at": "2026-11-01T16:31:00.993782Z", + "finished_at": "2026-11-01T16:31:50.962480Z" + }, + "agent_execution": { + "started_at": "2026-11-01T16:31:50.962661Z", + "finished_at": "2026-11-01T16:36:21.466005Z" + }, + "verifier": { + "started_at": "2026-11-01T16:36:21.700631Z", + "finished_at": "2026-11-01T16:39:02.082440Z" + } + }, + "agent_seconds": 270.503344, + "usage": { + "n_input_tokens": 137364, + "n_output_tokens": 11405, + "n_cache_tokens": 130048, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 17, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 17, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 17, + "assistant/message": 17, + "tool/call": 16, + "tool/result": 16, + "step/end": 17, + "turn/end": 1 + }, + "event_count": 110, + "tool_calls": 16, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "exposure_candidates": [], + "historical_pi": { + "native_reward": 0.0, + "analytical_score": 0.0, + "classification": "failed", + "reference_exposure": false, + "agent_seconds": 317.974853 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-query-optimize-cache-20261101.json": "fb1e727669d4c01a13753ae5122f3c8b1d433c221b32e0c317bb31d1e7fac9fa", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/result.json": "2a08fce9a87f184742e280a3004e3e7966296f5933ac72eabf2ec19764dc428b", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/manifest.json": "970c1ccb0b7c324cee5eae9052eac48e3f15196465d48d61e947046cb2f6c2fd", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101-input/config.json": "dc9b57cc66a6b6139452383e8ed9fb740875394ae14f20cbd5cc2dc8fabeebc7", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/agent/dsh-events.jsonl": "5d116300a3fde02997933ca5c7c771956a34ad4dad749f0c4b5919009e3caddb", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/test-stdout.txt": "5fdf69526b3cc8e8050cfe985df4b074b2ad9d569c5436eb4d3ddd3d2af9e572", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-query-optimize-cache-20261101/query-optimize__AEYyxag/verifier/retry-summary.json": "af00a7df4a76ece44d098028d1574868609704b24020c7a5f20670f4c8f9f953" + }, + "original_attempt": { + "job_name": "dsh-dev34-query-optimize-20261030", + "trial": "query-optimize__E5dWNng", + "native_reward": 0.0, + "native_exception_type": "NonZeroAgentExitCodeError", + "analytical_classification": "unscored_execution_error", + "analytical_score": null, + "diagnosis": "SDK terminated with finish_reason=error and a TRANSPORT error after six attempts of one provider request (five native request retries). Native NonZeroAgentExitCodeError retained. Official tests still ran (2 passed, 4 failed, missing SQL output), but the API-interrupted attempt is analytically unscored. No whole-trial retry." + }, + "diagnosis": "Completed normally; all six official SQL correctness, database-preservation, runtime, output-equivalence, single-query, and size checks passed. The prior terminal API error did not recur.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "public_task_dataset_download": false, + "environment_inspection": false + }, + "formal_dependency_profile": { + "profile": "query-optimize-v1", + "status": "passed" + } + }, + { + "task": "terminal-bench/torch-pipeline-parallelism", + "task_ref": "sha256:db605337c749a872cea7b5b413429b3915bb4c3efe0f7875f0c46ce81bd8c4fb", + "job_name": "dsh-sup6-torch-pipeline-parallelism-cache-20261101", + "trial": "torch-pipeline-parallelism__XRHZ5dL", + "arm": "deepseek", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "native_root": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL", + "record": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-pipeline-parallelism-cache-20261101.json", + "native_reward": 0.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 0.0 + }, + "analytical_classification": "failed", + "analytical_score": 0.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 2, + "failed": 2, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 2, + "failed": 1, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793552000.539176, + "stop": 1793552094.8557832 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 148.00501881899982 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-11-01T16:39:39.443791Z", + "finished_at": "2026-11-01T16:39:41.797007Z" + }, + "agent_setup": { + "started_at": "2026-11-01T16:39:41.797068Z", + "finished_at": "2026-11-01T16:41:55.045762Z" + }, + "agent_execution": { + "started_at": "2026-11-01T16:41:55.045883Z", + "finished_at": "2026-11-01T16:52:27.402287Z" + }, + "verifier": { + "started_at": "2026-11-01T16:52:27.635102Z", + "finished_at": "2026-11-01T16:54:55.881962Z" + } + }, + "agent_seconds": 632.356404, + "usage": { + "n_input_tokens": 444138, + "n_output_tokens": 25905, + "n_cache_tokens": 429056, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 31, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 31, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 31, + "assistant/message": 31, + "tool/call": 30, + "tool/result": 30, + "step/end": 31, + "turn/end": 1 + }, + "event_count": 194, + "tool_calls": 30, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "exposure_candidates": [ + "call_00_azBMWCwtE55WZURQIVO71536" + ], + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": true, + "agent_seconds": 562.196283 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-pipeline-parallelism-cache-20261101.json": "e2ff45301e23a775b6e044d69be6c4e62c3f9304769381512d5fce3b80ff9cce", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/result.json": "7cb7906719d93c44ea780360553ecbf746b9d997a2f3c611b5c36f9d98c91f32", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/manifest.json": "986021e2750ffa1f29629d8ef3188b555535460ab5ea7769af0e6d7bb6bf0a69", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101-input/config.json": "2e47fe00a10c5ceeb0ab4d649e17751289dd24ed4654f4f212a1874a5fdde23d", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/agent/dsh-events.jsonl": "388f8201b50db765149e4e93ec88cbc2e7814665e6c920a93c95df71b18b3fc6", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/test-stdout.txt": "69cb6842eab805f1b2699abb5f51242c4a4fdc207b216668926abb7974645d04", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-pipeline-parallelism-cache-20261101/torch-pipeline-parallelism__XRHZ5dL/verifier/retry-summary.json": "7443ea5e207b41cbcf0d965fe4ce34cba1e1aff258986ccd71d6e333754953fc" + }, + "original_attempt": { + "job_name": "dsh-dev34-torch-pipeline-parallelism-20261030", + "trial": "torch-pipeline-parallelism__zTzvuKq", + "native_reward": 0.0, + "native_exception_type": "AgentTimeoutError", + "analytical_classification": "unscored_verifier_error", + "analytical_score": null, + "diagnosis": "Native 900-second agent timeout was confirmed with 8 stopped processes. Official verifier then failed downloading nvidia-cusolver-cu12==11.7.1.2 required by torch==2.7.0, before pytest began. Dependency error makes the raw zero analytically null; native timeout retained." + }, + "diagnosis": "Completed normally. Existence and no-hooks checks passed; both world-size cases failed backward-value comparisons at microbatch 0. World size 1: rank 0 lm_head.bwd max difference 0.04166729375720024. World size 2: rank 0 model.layers.1.bwd max difference 0.01778167486190796. The worker exceptions carry these official assertions, not verifier infrastructure errors.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "public_task_dataset_download": false, + "environment_inspection": true + }, + "formal_dependency_profile": { + "profile": "torch-pipeline-parallelism-v1", + "status": "passed" + } + }, + { + "task": "terminal-bench/torch-tensor-parallelism", + "task_ref": "sha256:f32ce74a5aeb6638480247ab799fe46127bbee631acdd0921b0f394ec49b3684", + "job_name": "dsh-sup6-torch-tensor-parallelism-cache-20261101", + "trial": "torch-tensor-parallelism__kxNkcYR", + "arm": "deepseek", + "source_ref": "eff073b1601435305155cad92b6ee9e2250fbbdc", + "native_root": "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR", + "record": "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-tensor-parallelism-cache-20261101.json", + "native_reward": 1.0, + "native_exception_type": null, + "standard_summary_status": "scored", + "standard_summary_rewards": { + "reward": 1.0 + }, + "analytical_classification": "passed", + "analytical_score": 1.0, + "boundary_status": "confirmed", + "process_supervision": { + "status": "completed", + "return_code": 0 + }, + "sdk_finish_reason": "completed", + "pytest": { + "passed": 13, + "failed": 0, + "errors": 0, + "skipped": 0, + "xfailed": 0, + "xpassed": 0 + }, + "ctrf_summary": { + "tests": 3, + "passed": 3, + "failed": 0, + "skipped": 0, + "pending": 0, + "other": 0, + "start": 1793551442.7061565, + "stop": 1793551479.0017586 + }, + "verifier_attempts": { + "timeout_sec": 900.0, + "max_attempts": 3, + "retry_dependency_errors": false, + "retry_collection_errors": false, + "attempts": [ + { + "attempt": 1, + "status": "completed", + "return_code": 0, + "duration_seconds": 108.27482998300547 + } + ] + }, + "timing": { + "environment_setup": { + "started_at": "2026-11-01T16:39:39.419370Z", + "finished_at": "2026-11-01T16:39:41.760490Z" + }, + "agent_setup": { + "started_at": "2026-11-01T16:39:41.760533Z", + "finished_at": "2026-11-01T16:41:37.410735Z" + }, + "agent_execution": { + "started_at": "2026-11-01T16:41:37.410887Z", + "finished_at": "2026-11-01T16:42:50.780882Z" + }, + "verifier": { + "started_at": "2026-11-01T16:42:51.025511Z", + "finished_at": "2026-11-01T16:44:39.533328Z" + } + }, + "agent_seconds": 73.369995, + "usage": { + "n_input_tokens": 27764, + "n_output_tokens": 9845, + "n_cache_tokens": 25088, + "cost_usd": null, + "raw_deepseek_metadata": { + "profile": "sdk-minimal", + "source_commit": "f3803c96974ad4a94195fb97e5c637e9c2a3afe3", + "usage_records": 5, + "missing_usage_records": 0, + "malformed_event_records": 0, + "usage_is_partial": false, + "cost_is_unknown": true + } + }, + "operational_metadata": { + "event_types": { + "agent/inbox/spliced": 2, + "turn/start": 1, + "step/start": 5, + "system/message": 1, + "user/message": 2, + "request/header": 1, + "request/context": 1, + "session/title": 1, + "session-log-deepseek/delivery-accepted": 5, + "assistant/message": 5, + "tool/call": 4, + "tool/result": 4, + "step/end": 5, + "turn/end": 1 + }, + "event_count": 38, + "tool_calls": 4, + "events_after_agent_end": 0, + "events_at_or_after_verifier_start": 0 + }, + "exposure_candidates": [], + "historical_pi": { + "native_reward": 1.0, + "analytical_score": 1.0, + "classification": "passed", + "reference_exposure": false, + "agent_seconds": 90.783629 + }, + "evidence_sha256": { + "jobs/deepseek-harness-dev6-supplement-20261101-v2-orchestration/dsh-sup6-torch-tensor-parallelism-cache-20261101.json": "8f0d3027d7c1be04eb4453025fa7638dc25f8be7d16f8db8bc55573868673333", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/result.json": "c4b2c2af77c402de31017564e29740814d413752b6d306253fa90a29b6bc5360", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/manifest.json": "a0a307644bff0bc52ff2e0aa10d2f2cbc20ed9932cf5ec610d1a1992ec34b35a", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101-input/config.json": "419a58bbc8b1b219ebebf279ee1d5fc0f253880d89c3ec6d45f2cdbf5ce71b91", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/process-supervision.json": "c86e08060391f52429e8b442add8a03983fa0caf8f19eb62ad580208cc0bd7af", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/agent/dsh-events.jsonl": "d1f6f569271f12e30e9133598d27d0586229879675c85aba01ce729f87c9bfa8", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/test-stdout.txt": "92f8c999585b93e3575786304e330d327b043e24eec9b2080df45617ea782b88", + "jobs/deepseek-harness-supplement-source/jobs/dsh-sup6-torch-tensor-parallelism-cache-20261101/torch-tensor-parallelism__kxNkcYR/verifier/retry-summary.json": "2d71d0cb1ba3c6a823496a68f073fc3b0806c73b323867d1e059d2c2c7b615e3" + }, + "original_attempt": { + "job_name": "dsh-dev34-torch-tensor-parallelism-20261030", + "trial": "torch-tensor-parallelism__ZVRUdrm", + "native_reward": 0.0, + "native_exception_type": null, + "analytical_classification": "unscored_verifier_error", + "analytical_score": null, + "diagnosis": "First verifier attempt hit the native 900-second timeout. The second attempt failed downloading torch==2.7.0 with UV_HTTP_TIMEOUT=30 before pytest began. Raw zero retained; analytical score null." + }, + "diagnosis": "Completed normally; all 13 official interface and parameterized column/row-parallel linear checks passed.", + "audit_flags": { + "service_lifecycle_caveat": false, + "external_benchmark_material_observed": false, + "provider_key_exact_match_in_frozen_files": false, + "public_task_dataset_download": false, + "environment_inspection": false + }, + "formal_dependency_profile": { + "profile": "torch-tensor-parallelism-v1", + "status": "passed" + } + } + ] +} From 18aecddb15ca5885eeed71a20ffbdd6fe657f6de Mon Sep 17 00:00:00 2001 From: minixalpha Date: Mon, 2 Nov 2026 06:10:29 -0800 Subject: [PATCH 15/15] docs: record the deepseek harness benchmark workflow --- docs/changelogs/0.10.x.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/docs/changelogs/0.10.x.md b/docs/changelogs/0.10.x.md index cb7ecf6..24d7511 100644 --- a/docs/changelogs/0.10.x.md +++ b/docs/changelogs/0.10.x.md @@ -11,6 +11,9 @@ All notable changes in the **0.10.x** release series are documented here. - Add pinned Pi agent baselines to the standard Harbor workflow, with offline model-registration checks, verified image caching, process supervision before grading, and explicit partial usage and cost accounting. +- Add official DeepSeek Harness `sdk-minimal` baselines to the standard Harbor + workflow, with pinned offline SDK/runtime artifacts, verified PyPI wheel + caching, reviewed dependency profiles, and bilingual audited comparison reports. ### Changed - Clarify headless task contracts, output verification, and permitted references.