From ab4b379c1a209f5bf8446e97266dd09590b5f70e Mon Sep 17 00:00:00 2001 From: Rudrasinh Nimeshkumar Ravalji <52209277+rudra717@users.noreply.github.com> Date: Tue, 7 Apr 2026 20:59:01 -0700 Subject: [PATCH] fix(model-registry): add supports_reasoning to all gpt-oss model entries (#25160) gpt-oss is a reasoning model, but 16 provider entries were missing the supports_reasoning flag. This caused litellm to reject reasoning parameters (e.g. reasoning_effort) for these providers, raising UnsupportedParamsError. Updated providers: azure_ai, databricks, deepinfra, lemonade, ollama, replicate, together_ai, wandb, watsonx. Safeguard variants (gpt-oss-safeguard-*) are excluded as they are non-reasoning variants. Fixes BerriAI/litellm#25132 Co-authored-by: Claude --- ...odel_prices_and_context_window_backup.json | 48 ++++++++++++------- 1 file changed, 32 insertions(+), 16 deletions(-) diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index d781c91992d..6ebaecf049c 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -1875,7 +1875,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "supports_reasoning": true }, "azure_ai/model_router": { "input_cost_per_token": 1.4e-07, @@ -10397,7 +10398,8 @@ "mode": "chat", "output_cost_per_token": 5.9997e-07, "output_dbu_cost_per_token": 8.571e-06, - "source": "https://www.databricks.com/product/pricing/foundation-model-serving" + "source": "https://www.databricks.com/product/pricing/foundation-model-serving", + "supports_reasoning": true }, "databricks/databricks-gpt-oss-20b": { "input_cost_per_token": 7e-08, @@ -10412,7 +10414,8 @@ "mode": "chat", "output_cost_per_token": 3.0001999999999996e-07, "output_dbu_cost_per_token": 4.285999999999999e-06, - "source": "https://www.databricks.com/product/pricing/foundation-model-serving" + "source": "https://www.databricks.com/product/pricing/foundation-model-serving", + "supports_reasoning": true }, "databricks/databricks-gte-large-en": { "input_cost_per_token": 1.2999000000000001e-07, @@ -11794,7 +11797,8 @@ "litellm_provider": "deepinfra", "mode": "chat", "supports_tool_choice": true, - "supports_function_calling": true + "supports_function_calling": true, + "supports_reasoning": true }, "deepinfra/openai/gpt-oss-20b": { "max_tokens": 131072, @@ -11805,7 +11809,8 @@ "litellm_provider": "deepinfra", "mode": "chat", "supports_tool_choice": true, - "supports_function_calling": true + "supports_function_calling": true, + "supports_reasoning": true }, "deepinfra/zai-org/GLM-4.5": { "max_tokens": 131072, @@ -20088,7 +20093,8 @@ "output_cost_per_token": 0, "supports_function_calling": true, "supports_response_schema": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "supports_reasoning": true }, "lemonade/gpt-oss-120b-mxfp-GGUF": { "input_cost_per_token": 0, @@ -20100,7 +20106,8 @@ "output_cost_per_token": 0, "supports_function_calling": true, "supports_response_schema": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "supports_reasoning": true }, "lemonade/Gemma-3-4b-it-GGUF": { "input_cost_per_token": 0, @@ -24345,7 +24352,8 @@ "max_tokens": 131072, "mode": "chat", "output_cost_per_token": 0.0, - "supports_function_calling": true + "supports_function_calling": true, + "supports_reasoning": true }, "ollama/gpt-oss:20b-cloud": { "input_cost_per_token": 0.0, @@ -24355,7 +24363,8 @@ "max_tokens": 131072, "mode": "chat", "output_cost_per_token": 0.0, - "supports_function_calling": true + "supports_function_calling": true, + "supports_reasoning": true }, "ollama/internlm2_5-20b-chat": { "input_cost_per_token": 0.0, @@ -26979,7 +26988,8 @@ "litellm_provider": "replicate", "mode": "chat", "supports_function_calling": true, - "supports_system_messages": true + "supports_system_messages": true, + "supports_reasoning": true }, "replicate/anthropic/claude-4.5-haiku": { "input_cost_per_token": 1e-06, @@ -27221,7 +27231,8 @@ "litellm_provider": "replicate", "mode": "chat", "supports_function_calling": true, - "supports_system_messages": true + "supports_system_messages": true, + "supports_reasoning": true }, "replicate/deepseek-ai/deepseek-v3.1": { "input_cost_per_token": 6.72e-07, @@ -28558,7 +28569,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "supports_reasoning": true }, "together_ai/openai/gpt-oss-20b": { "input_cost_per_token": 5e-08, @@ -28570,7 +28582,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true, - "supports_tool_choice": true + "supports_tool_choice": true, + "supports_reasoning": true }, "together_ai/togethercomputer/CodeLlama-34b-Instruct": { "litellm_provider": "together_ai", @@ -32259,7 +32272,8 @@ "input_cost_per_token": 0.015, "output_cost_per_token": 0.06, "litellm_provider": "wandb", - "mode": "chat" + "mode": "chat", + "supports_reasoning": true }, "wandb/openai/gpt-oss-20b": { "max_tokens": 131072, @@ -32268,7 +32282,8 @@ "input_cost_per_token": 0.005, "output_cost_per_token": 0.02, "litellm_provider": "wandb", - "mode": "chat" + "mode": "chat", + "supports_reasoning": true }, "wandb/zai-org/GLM-4.5": { "max_tokens": 131072, @@ -32712,7 +32727,8 @@ "mode": "chat", "supports_function_calling": false, "supports_parallel_function_calling": false, - "supports_vision": false + "supports_vision": false, + "supports_reasoning": true }, "watsonx/sdaia/allam-1-13b-instruct": { "max_tokens": 8192,