Workflow runs · lm-sys/FastChat · GitHub

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows

95 workflow runs

95 workflow runs

fix: use multi-GPU RWKV strategy from visible CUDA devices Python package #4113: Pull request #3882 synchronize by Chessing234

40s Chessing234:fix/rwkv-multi-gpu-strategy

Chessing234:fix/rwkv-multi-gpu-strategy

40s

fix: detect partial stop sequences by output suffix Python package #4112: Pull request #3883 synchronize by Chessing234

49s Chessing234:fix/is-partial-stop-suffix-check

Chessing234:fix/is-partial-stop-suffix-check

49s

fix: allow separate input/output budgets for T5 in context check Python package #4111: Pull request #3885 synchronize by Chessing234

40s Chessing234:fix/t5-encoder-decoder-context-check

Chessing234:fix/t5-encoder-decoder-context-check

40s

Fix arena moderation skipping right model conversation history Python package #4110: Pull request #3889 synchronize by Chessing234

50s Chessing234:fix/arena-moderation-missing-model-b-history

Chessing234:fix/arena-moderation-missing-model-b-history

50s

fix: remove off-by-one in inference prompt truncation Python package #4109: Pull request #3890 synchronize by Chessing234

41s Chessing234:fix/inference-truncate-max-src-len

Chessing234:fix/inference-truncate-max-src-len

41s

fix: check partial stop on cumulative sglang output Python package #4108: Pull request #3891 synchronize by Chessing234

50s Chessing234:fix/sglang-partial-stop-cumulative-output

Chessing234:fix/sglang-partial-stop-cumulative-output

50s

Fix typos in multi-turn judge prompt templates Python package #4107: Pull request #3892 synchronize by Chessing234

43s Chessing234:fix/judge-prompt-typos

Chessing234:fix/judge-prompt-typos

43s

fix(webui): confirm before clearing chat history Python package #4106: Pull request #3893 synchronize by Chessing234

46s Chessing234:fix/3881-clear-history-confirmation

Chessing234:fix/3881-clear-history-confirmation

46s

fix: block SSRF to private hosts in load_image Python package #4105: Pull request #3894 synchronize by Chessing234

44s Chessing234:fix/3873-load-image-ssrf

Chessing234:fix/3873-load-image-ssrf

44s

fix: map uvicorn stderr prefixes to correct log levels Python package #4104: Pull request #3895 synchronize by Chessing234

39s Chessing234:fix/1292-stderr-log-level

Chessing234:fix/1292-stderr-log-level

39s

fix: model worker --debug flag parsing Python package #4103: Pull request #3896 synchronize by Chessing234

38s Chessing234:fix/model-worker-debug-flag

Chessing234:fix/model-worker-debug-flag

38s

fix: drop single-$ latex delimiters so currency like 25$ renders Python package #4102: Pull request #3905 synchronize by Chessing234

40s Chessing234:fix/3904-currency-breaks-latex

Chessing234:fix/3904-currency-breaks-latex

40s

fix: reject local filesystem paths in load_image Python package #4101: Pull request #3906 synchronize by Chessing234

48s Chessing234:fix/3873-reject-local-image-paths

Chessing234:fix/3873-reject-local-image-paths

48s

fix: tolerate rope_scaling without factor key Python package #4100: Pull request #3907 synchronize by Chessing234

45s Chessing234:fix/3470-rope-scaling-factor-key

Chessing234:fix/3470-rope-scaling-factor-key

45s

fix: CUDA OOM handling on older torch Python package #4099: Pull request #3908 synchronize by Chessing234

38s Chessing234:fix/1627-cuda-oom-old-torch

Chessing234:fix/1627-cuda-oom-old-torch

38s

fix: clean tokenization spaces in huggingface_api Python package #4098: Pull request #3909 synchronize by Chessing234

52s Chessing234:fix/3795-huggingface-api-clean-tokenization

Chessing234:fix/3795-huggingface-api-clean-tokenization

52s

fix: register workers with matching address in launch_all_serve Python package #4097: Pull request #3919 synchronize by Chessing234

42s Chessing234:fix/launch-all-serve-worker-address

Chessing234:fix/launch-all-serve-worker-address

42s

docs: note OpenAI client base_url for multi-model gateways Python package #4096: Pull request #3920 opened by seven7763

44s seven7763:docs/openai-compatible-gateway-example

seven7763:docs/openai-compatible-gateway-example

44s

fix: register workers with matching address in launch_all_serve Python package #4095: Pull request #3919 opened by Chessing234

39s Chessing234:fix/launch-all-serve-worker-address

Chessing234:fix/launch-all-serve-worker-address

39s

fix: model worker --debug flag parsing Python package #4094: Pull request #3896 synchronize by Chessing234

50s Chessing234:fix/model-worker-debug-flag

Chessing234:fix/model-worker-debug-flag

50s

fix: use multi-GPU RWKV strategy from visible CUDA devices Python package #4093: Pull request #3882 synchronize by Chessing234

47s Chessing234:fix/rwkv-multi-gpu-strategy

Chessing234:fix/rwkv-multi-gpu-strategy

47s

fix: clean tokenization spaces in huggingface_api Python package #4092: Pull request #3909 opened by Chessing234

38s Chessing234:fix/3795-huggingface-api-clean-tokenization

Chessing234:fix/3795-huggingface-api-clean-tokenization

38s

fix: CUDA OOM handling on older torch Python package #4091: Pull request #3908 opened by Chessing234

46s Chessing234:fix/1627-cuda-oom-old-torch

Chessing234:fix/1627-cuda-oom-old-torch

46s

fix: tolerate rope_scaling without factor key Python package #4090: Pull request #3907 opened by Chessing234

37s Chessing234:fix/3470-rope-scaling-factor-key

Chessing234:fix/3470-rope-scaling-factor-key

37s

fix: reject local filesystem paths in load_image Python package #4089: Pull request #3906 opened by Chessing234

44s Chessing234:fix/3873-reject-local-image-paths

Chessing234:fix/3873-reject-local-image-paths

44s