From 3ae5ecc79cefe099d60de215955f553b963600ec Mon Sep 17 00:00:00 2001 From: C Naresh Date: Wed, 14 Oct 2026 00:07:04 +0530 Subject: [PATCH 1/2] feat(google): add gemini-4-argon to Gemini Developer API Gemini 4 Argon is available on the Gemini Developer API (generativelanguage), so add it alongside the existing Vertex entry. Config: - 65,536 max output tokens (per-request synchronous cap; 1M is only reachable via long-decode continuation tokens across multiple requests) - reasoning_effort low/medium/high instead of thinking.budget_tokens: Argon returns HTTP 400 for thinking_budget and only accepts thinking levels - removeParams temperature/top_p/top_k: Argon ignores sampling parameters - cache_read only, no cache_write: implicit context caching only Pricing (promotional rate in effect through 2026-11-30), $/1M -> cents/token: - input $2 -> 0.0002, output $10 -> 0.001, cached input $0.10 -> 0.00001 - batch input $1 -> 0.0001, batch output $5 -> 0.0005 - grounding $14/1k queries -> 1.4 Keyed as -lte-128k/-gt-128k because the google pricing handler always appends a context-tier suffix; Argon is not context-tiered so both are identical. Source: https://ai.google.dev/gemini-api/docs/pricing https://ai.google.dev/gemini-api/docs/models/gemini-4-argon --- general/google.json | 51 +++++++++++++++++++++++++++++++++++++ pricing/google.json | 62 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 113 insertions(+) diff --git a/general/google.json b/general/google.json index 4dbe50fe2..5423cd53a 100644 --- a/general/google.json +++ b/general/google.json @@ -2043,6 +2043,57 @@ ], "type": { "primary": "chat", "supported": ["image", "pdf", "doc", "tools", "audio"] } }, + "gemini-4-argon": { + "params": [ + { "key": "max_tokens", "maxValue": 65536 }, + { + "key": "response_format", + "defaultValue": null, + "options": [ + { + "value": null, + "name": "Text" + }, + { + "value": "json_object", + "name": "JSON Object", + "schema": { + "type": "object", + "properties": { + "type": { "type": "string", "value": "json_object" } + } + } + }, + { + "value": "json_schema", + "name": "JSON Schema", + "schema": { + "type": "object", + "properties": { + "type": { "type": "string", "value": "json_schema" }, + "json_schema": { "type": "object" } + } + }, + "params": { "key": "json_schema", "defaultValue": null, "type": "json", "skipValues": [null] } + } + ], + "skipValues": [null], + "type": "string" + }, + { + "key": "reasoning_effort", + "defaultValue": "medium", + "options": [ + { "value": "low", "name": "Low" }, + { "value": "medium", "name": "Medium" }, + { "value": "high", "name": "High" } + ], + "type": "string" + } + ], + "type": { "primary": "chat", "supported": ["image", "pdf", "doc", "tools", "audio"] }, + "removeParams": ["temperature", "top_p", "top_k"] + }, "gemini-3.1-flash-image-preview": { "type": { "primary": "image" diff --git a/pricing/google.json b/pricing/google.json index 4ba1a7433..ee3402d97 100644 --- a/pricing/google.json +++ b/pricing/google.json @@ -3948,5 +3948,67 @@ } } } + }, + "gemini-4-argon-lte-128k": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0002 + }, + "response_token": { + "price": 0.001 + }, + "cache_read_input_token": { + "price": 0.00001 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + } + } + }, + "batch_config": { + "request_token": { + "price": 0.0001 + }, + "response_token": { + "price": 0.0005 + } + } + } + }, + "gemini-4-argon-gt-128k": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0002 + }, + "response_token": { + "price": 0.001 + }, + "cache_read_input_token": { + "price": 0.00001 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + } + } + }, + "batch_config": { + "request_token": { + "price": 0.0001 + }, + "response_token": { + "price": 0.0005 + } + } + } } } From f61d65b2d6177821752b162fa7d3b47fb44b8ca4 Mon Sep 17 00:00:00 2001 From: C Naresh Date: Wed, 14 Oct 2026 00:07:18 +0530 Subject: [PATCH 2/2] fix(vertex-ai): complete gemini-4-argon pricing and correct its params The initial gemini-4-argon entry only carried a single flat rate, so non-global regions and the priority/flex/batch execution modes were all billed at the global standard rate. Fill in the region x execution-mode matrix that every other Gemini model in this file already has. Rates (promotional, in effect through 2026-11-30), $/1M -> cents/token: global non-global (default) standard $2 / $0.10 / $10 $2.2 / $0.11 / $11 batch/flex $1 / $0.05 / $5 $1.1 / $0.055 / $5.5 priority $3.6 / $0.18 / $18 $3.96 / $0.198 / $19.8 Also corrects gemini-4-argon-cyber-permissive, which had been given the promotional rate. The pricing page footnote states the promotion does not apply to Cyber Permissive, so it takes the list rate ($4 in / $20 out / $0.20 cached, +10% non-global) - previously undercharged by 50%. Params: Argon returns HTTP 400 when thinking_budget is set and only accepts thinking levels, so thinking.budget_tokens is replaced with reasoning_effort (low/medium/high, default medium), which the gateway already maps to thinkingConfig.thinkingLevel. temperature/top_p are removed since Argon ignores sampling parameters. Source: https://cloud.google.com/vertex-ai/generative-ai/pricing https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-argon --- general/vertex-ai.json | 25 +- pricing/vertex-ai.json | 512 ++++++++++++++++++++++++++++++++++++++++- 2 files changed, 519 insertions(+), 18 deletions(-) diff --git a/general/vertex-ai.json b/general/vertex-ai.json index b6210bb24..4776e2e2e 100644 --- a/general/vertex-ai.json +++ b/general/vertex-ai.json @@ -2504,23 +2504,18 @@ "type": "string" }, { - "key": "thinking", - "properties": { - "type": { - "type": "string", - "enum": ["enabled"] - }, - "budget_tokens": { - "type": "number", - "minValue": 128, - "maxValue": 32768 - } - }, - "defaultValue": null, - "skipValues": [null] + "key": "reasoning_effort", + "defaultValue": "medium", + "options": [ + { "value": "low", "name": "Low" }, + { "value": "medium", "name": "Medium" }, + { "value": "high", "name": "High" } + ], + "type": "string" } ], - "type": { "primary": "chat", "supported": ["image", "pdf", "doc", "tools", "audio"] } + "type": { "primary": "chat", "supported": ["image", "pdf", "doc", "tools", "audio"] }, + "removeParams": ["temperature", "top_p"] }, "gemini-3.1-pro-preview-customtools": { "params": [ diff --git a/pricing/vertex-ai.json b/pricing/vertex-ai.json index 0e83dff88..ab6dbb421 100644 --- a/pricing/vertex-ai.json +++ b/pricing/vertex-ai.json @@ -17113,13 +17113,13 @@ "pricing_config": { "pay_as_you_go": { "request_token": { - "price": 0.0002 + "price": 0.0004 }, "response_token": { - "price": 0.001 + "price": 0.002 }, "cache_read_input_token": { - "price": 0.00001 + "price": 0.00002 }, "additional_units": { "web_search": { @@ -17135,9 +17135,262 @@ "price": 1.4 } } + }, + "batch_config": { + "request_token": { + "price": 0.0002 + }, + "response_token": { + "price": 0.001 + } + } + }, + "custom_pricing": { + "regions": { + "global": { + "execution_modes": { + "standard": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0004 + }, + "cache_read_input_token": { + "price": 0.00002 + }, + "response_token": { + "price": 0.002 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "batch": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0002 + }, + "cache_read_input_token": { + "price": 0.00001 + }, + "response_token": { + "price": 0.001 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "flex": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0002 + }, + "cache_read_input_token": { + "price": 0.00001 + }, + "response_token": { + "price": 0.001 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "priority": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00072 + }, + "cache_read_input_token": { + "price": 0.000036 + }, + "response_token": { + "price": 0.0036 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + } + } + }, + "default": { + "execution_modes": { + "standard": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00044 + }, + "cache_read_input_token": { + "price": 0.000022 + }, + "response_token": { + "price": 0.0022 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "batch": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00022 + }, + "cache_read_input_token": { + "price": 0.000011 + }, + "response_token": { + "price": 0.0011 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "flex": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00022 + }, + "cache_read_input_token": { + "price": 0.000011 + }, + "response_token": { + "price": 0.0011 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "priority": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.000792 + }, + "cache_read_input_token": { + "price": 0.0000396 + }, + "response_token": { + "price": 0.00396 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + } + } + } } } }, + "gemini-4-argon": { "pricing_config": { "pay_as_you_go": { @@ -17164,9 +17417,262 @@ "price": 1.4 } } + }, + "batch_config": { + "request_token": { + "price": 0.0001 + }, + "response_token": { + "price": 0.0005 + } + } + }, + "custom_pricing": { + "regions": { + "global": { + "execution_modes": { + "standard": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0002 + }, + "cache_read_input_token": { + "price": 0.00001 + }, + "response_token": { + "price": 0.001 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "batch": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0001 + }, + "cache_read_input_token": { + "price": 0.000005 + }, + "response_token": { + "price": 0.0005 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "flex": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.0001 + }, + "cache_read_input_token": { + "price": 0.000005 + }, + "response_token": { + "price": 0.0005 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "priority": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00036 + }, + "cache_read_input_token": { + "price": 0.000018 + }, + "response_token": { + "price": 0.0018 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + } + } + }, + "default": { + "execution_modes": { + "standard": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00022 + }, + "cache_read_input_token": { + "price": 0.000011 + }, + "response_token": { + "price": 0.0011 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "batch": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00011 + }, + "cache_read_input_token": { + "price": 0.0000055 + }, + "response_token": { + "price": 0.00055 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "flex": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.00011 + }, + "cache_read_input_token": { + "price": 0.0000055 + }, + "response_token": { + "price": 0.00055 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + }, + "priority": { + "pricing_config": { + "pay_as_you_go": { + "request_token": { + "price": 0.000396 + }, + "cache_read_input_token": { + "price": 0.0000198 + }, + "response_token": { + "price": 0.00198 + }, + "additional_units": { + "web_search": { + "price": 1.4 + }, + "search": { + "price": 1.4 + }, + "enterprise_web_search": { + "price": 1.4 + }, + "maps": { + "price": 1.4 + } + } + } + } + } + } + } } } }, + "gemini-4": { "pricing_config": { "pay_as_you_go": {