From c060b59e2cb21c551b339172843140d01bc5c2f7 Mon Sep 17 00:00:00 2001 From: Sameer Kankute Date: Fri, 6 Mar 2026 18:16:29 +0530 Subject: [PATCH] Merge pull request #22953 from BerriAI/litellm_gpt5_reasoning_model_map feat(gpt-5): Add supports_none_reasoning_effort and supports_xhigh_reasoning_effort to model cost map --- .../llms/azure/chat/gpt_5_transformation.py | 16 +- .../llms/openai/chat/gpt_5_transformation.py | 127 +++++--- ...odel_prices_and_context_window_backup.json | 296 ++++++++++++++++-- litellm/types/utils.py | 2 + model_prices_and_context_window.json | 296 ++++++++++++++++-- .../llms/openai/test_gpt5_transformation.py | 263 +++++++++++++++- 6 files changed, 890 insertions(+), 110 deletions(-) diff --git a/litellm/llms/azure/chat/gpt_5_transformation.py b/litellm/llms/azure/chat/gpt_5_transformation.py index eeb55911ecf..ae782fdce2e 100644 --- a/litellm/llms/azure/chat/gpt_5_transformation.py +++ b/litellm/llms/azure/chat/gpt_5_transformation.py @@ -43,8 +43,12 @@ class AzureOpenAIGPT5Config(AzureOpenAIConfig, OpenAIGPT5Config): if "tool_choice" not in params: params.append("tool_choice") - # Only gpt-5.2 has been verified to support logprobs on Azure - if self.is_model_gpt_5_2_model(model): + # Only gpt-5.2+ has been verified to support logprobs on Azure. + # The base OpenAI class includes logprobs for gpt-5.1+, but Azure + # hasn't verified support for gpt-5.1, so remove them unless gpt-5.2/5.4+. + if self._supports_reasoning_effort_level(model, "none") and not self.is_model_gpt_5_2_model(model): + params = [p for p in params if p not in ["logprobs", "top_logprobs"]] + elif self.is_model_gpt_5_2_model(model): azure_supported_params = ["logprobs", "top_logprobs"] params.extend(azure_supported_params) @@ -65,9 +69,9 @@ class AzureOpenAIGPT5Config(AzureOpenAIConfig, OpenAIGPT5Config): # gpt-5.1 supports reasoning_effort='none', but other gpt-5 models don't # See: https://learn.microsoft.com/en-us/azure/ai-foundry/openai/how-to/reasoning - is_gpt_5_1 = self.is_model_gpt_5_1_model(model) + supports_none = self._supports_reasoning_effort_level(model, "none") - if reasoning_effort_value == "none" and not is_gpt_5_1: + if reasoning_effort_value == "none" and not supports_none: if litellm.drop_params is True or ( drop_params is not None and drop_params is True ): @@ -97,8 +101,8 @@ class AzureOpenAIGPT5Config(AzureOpenAIConfig, OpenAIGPT5Config): drop_params=drop_params, ) - # Only drop reasoning_effort='none' for non-gpt-5.1 models - if result.get("reasoning_effort") == "none" and not is_gpt_5_1: + # Only drop reasoning_effort='none' for models that don't support it + if result.get("reasoning_effort") == "none" and not supports_none: result.pop("reasoning_effort") return result diff --git a/litellm/llms/openai/chat/gpt_5_transformation.py b/litellm/llms/openai/chat/gpt_5_transformation.py index 05c003c8b7a..358a94016ae 100644 --- a/litellm/llms/openai/chat/gpt_5_transformation.py +++ b/litellm/llms/openai/chat/gpt_5_transformation.py @@ -3,6 +3,7 @@ from typing import Optional import litellm +from litellm.utils import _supports_factory from .gpt_transformation import OpenAIGPTConfig @@ -23,43 +24,61 @@ class OpenAIGPT5Config(OpenAIGPTConfig): # Don't route it through GPT-5 reasoning-specific parameter restrictions. return "gpt-5" in model and "gpt-5-chat" not in model + @classmethod + def is_model_gpt_5_search_model(cls, model: str) -> bool: + """Check if the model is a GPT-5 search variant (e.g. gpt-5-search-api). + + Search-only models have a severely restricted parameter set compared to + regular GPT-5 models. They are identified by name convention (contain + both ``gpt-5`` and ``search``). Note: ``supports_web_search`` in model + info is a *different* concept — it indicates a model can *use* web + search as a tool, which many non-search-only models also support. + """ + return "gpt-5" in model and "search" in model + @classmethod def is_model_gpt_5_codex_model(cls, model: str) -> bool: """Check if the model is specifically a GPT-5 Codex variant.""" return "gpt-5-codex" in model - @classmethod - def is_model_gpt_5_1_codex_max_model(cls, model: str) -> bool: - """Check if the model is the gpt-5.1-codex-max variant.""" - model_name = model.split("/")[-1] # handle provider prefixes - return model_name == "gpt-5.1-codex-max" - - @classmethod - def is_model_gpt_5_1_model(cls, model: str) -> bool: - """Check if the model is a gpt-5.1 or gpt-5.2 chat variant. - - gpt-5.1/5.2 support temperature when reasoning_effort="none", - unlike base gpt-5 which only supports temperature=1. Excludes - pro variants which keep stricter knobs. - """ - model_name = model.split("/")[-1] - is_gpt_5_1 = model_name.startswith("gpt-5.1") - is_gpt_5_2 = model_name.startswith("gpt-5.2") and "pro" not in model_name - return is_gpt_5_1 or is_gpt_5_2 - - @classmethod - def is_model_gpt_5_2_pro_model(cls, model: str) -> bool: - """Check if the model is the gpt-5.2-pro snapshot/alias.""" - model_name = model.split("/")[-1] - return model_name.startswith("gpt-5.2-pro") - @classmethod def is_model_gpt_5_2_model(cls, model: str) -> bool: """Check if the model is a gpt-5.2 variant (including pro).""" model_name = model.split("/")[-1] - return model_name.startswith("gpt-5.2") + return model_name.startswith("gpt-5.2") or model_name.startswith("gpt-5.4") + + @classmethod + def _supports_reasoning_effort_level(cls, model: str, level: str) -> bool: + """Check if the model supports a specific reasoning_effort level. + + Looks up ``supports_{level}_reasoning_effort`` in the model map via + the shared ``_supports_factory`` helper. + Returns False for unknown models (safe fallback). + """ + return _supports_factory( + model=model, + custom_llm_provider=None, + key=f"supports_{level}_reasoning_effort", + ) def get_supported_openai_params(self, model: str) -> list: + if self.is_model_gpt_5_search_model(model): + return [ + "max_tokens", + "max_completion_tokens", + "stream", + "stream_options", + "web_search_options", + "service_tier", + "safety_identifier", + "response_format", + "user", + "store", + "verbosity", + "max_retries", + "extra_headers", + ] + from litellm.utils import supports_tool_choice base_gpt_series_params = super().get_supported_openai_params(model=model) @@ -69,14 +88,20 @@ class OpenAIGPT5Config(OpenAIGPTConfig): base_gpt_series_params.remove("tool_choice") non_supported_params = [ - "logprobs", - "top_p", "presence_penalty", "frequency_penalty", - "top_logprobs", "stop", + "logit_bias", + "modalities", + "prediction", + "audio", + "web_search_options", ] + # gpt-5.1/5.2 support logprobs, top_p, top_logprobs when reasoning_effort="none" + if not self._supports_reasoning_effort_level(model, "none"): + non_supported_params.extend(["logprobs", "top_p", "top_logprobs"]) + return [ param for param in base_gpt_series_params @@ -90,21 +115,30 @@ class OpenAIGPT5Config(OpenAIGPTConfig): model: str, drop_params: bool, ) -> dict: + if self.is_model_gpt_5_search_model(model): + if "max_tokens" in non_default_params: + optional_params["max_completion_tokens"] = non_default_params.pop( + "max_tokens" + ) + return super()._map_openai_params( + non_default_params=non_default_params, + optional_params=optional_params, + model=model, + drop_params=drop_params, + ) + reasoning_effort = ( non_default_params.get("reasoning_effort") or optional_params.get("reasoning_effort") ) if reasoning_effort is not None and reasoning_effort == "xhigh": - if not ( - self.is_model_gpt_5_1_codex_max_model(model) - or self.is_model_gpt_5_2_model(model) - ): + if not self._supports_reasoning_effort_level(model, "xhigh"): if litellm.drop_params or drop_params: non_default_params.pop("reasoning_effort", None) else: raise litellm.utils.UnsupportedParamsError( message=( - "reasoning_effort='xhigh' is only supported for gpt-5.1-codex-max and gpt-5.2 models." + "reasoning_effort='xhigh' is only supported for gpt-5.1-codex-max, gpt-5.2, and gpt-5.4+ models." ), status_code=400, ) @@ -118,13 +152,30 @@ class OpenAIGPT5Config(OpenAIGPTConfig): "max_tokens" ) + # gpt-5.1/5.2 support logprobs, top_p, top_logprobs only when reasoning_effort="none" + supports_none = self._supports_reasoning_effort_level(model, "none") + if supports_none: + sampling_params = ["logprobs", "top_logprobs", "top_p"] + has_sampling = any(p in non_default_params for p in sampling_params) + if has_sampling and reasoning_effort not in (None, "none"): + if litellm.drop_params or drop_params: + for p in sampling_params: + non_default_params.pop(p, None) + else: + raise litellm.utils.UnsupportedParamsError( + message=( + "gpt-5.1/5.2/5.4 only support logprobs, top_p, top_logprobs when " + "reasoning_effort='none'. Current reasoning_effort='{}'. " + "To drop unsupported params set `litellm.drop_params = True`" + ).format(reasoning_effort), + status_code=400, + ) + if "temperature" in non_default_params: temperature_value: Optional[float] = non_default_params.pop("temperature") if temperature_value is not None: - is_gpt_5_1 = self.is_model_gpt_5_1_model(model) - - # gpt-5.1 supports any temperature when reasoning_effort="none" (or not specified, as it defaults to "none") - if is_gpt_5_1 and (reasoning_effort == "none" or reasoning_effort is None): + # models supporting reasoning_effort="none" also support flexible temperature + if supports_none and (reasoning_effort == "none" or reasoning_effort is None): optional_params["temperature"] = temperature_value elif temperature_value == 1: optional_params["temperature"] = temperature_value diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index f52288ea72a..3bacdfddf2f 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -20113,7 +20113,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1": { "cache_read_input_token_cost": 1.25e-07, @@ -20149,7 +20151,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-2025-11-13": { "cache_read_input_token_cost": 1.25e-07, @@ -20185,7 +20189,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-chat-latest": { "cache_read_input_token_cost": 1.25e-07, @@ -20220,7 +20226,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": false, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": false }, "gpt-5.2": { "cache_read_input_token_cost": 1.75e-07, @@ -20257,7 +20265,10 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": true, + "supports_web_search": true }, "gpt-5.2-2025-12-11": { "cache_read_input_token_cost": 1.75e-07, @@ -20294,7 +20305,10 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": true, + "supports_web_search": true }, "gpt-5.2-chat-latest": { "cache_read_input_token_cost": 1.75e-07, @@ -20328,7 +20342,47 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false, + "supports_web_search": true + }, + "gpt-5.3-chat-latest": { + "cache_read_input_token_cost": 1.75e-07, + "cache_read_input_token_cost_priority": 3.5e-07, + "input_cost_per_token": 1.75e-06, + "input_cost_per_token_priority": 3.5e-06, + "litellm_provider": "openai", + "max_input_tokens": 128000, + "max_output_tokens": 16384, + "max_tokens": 16384, + "mode": "chat", + "output_cost_per_token": 1.4e-05, + "output_cost_per_token_priority": 2.8e-05, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false, + "supports_web_search": true }, "gpt-5.2-pro": { "input_cost_per_token": 2.1e-05, @@ -20359,7 +20413,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5.2-pro-2025-12-11": { "input_cost_per_token": 2.1e-05, @@ -20390,7 +20446,161 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true + }, + "gpt-5.4": { + "cache_read_input_token_cost": 2.5e-07, + "cache_read_input_token_cost_priority": 5e-07, + "input_cost_per_token": 2.5e-06, + "input_cost_per_token_priority": 5e-06, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.5e-05, + "output_cost_per_token_priority": 2.25e-05, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": true + }, + "gpt-5.4-2026-03-05": { + "cache_read_input_token_cost": 2.5e-07, + "cache_read_input_token_cost_priority": 5e-07, + "input_cost_per_token": 2.5e-06, + "input_cost_per_token_priority": 5e-06, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.5e-05, + "output_cost_per_token_priority": 2.25e-05, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true + }, + "gpt-5.4-pro": { + "cache_read_input_token_cost": 3e-06, + "cache_read_input_token_cost_priority": 6e-06, + "input_cost_per_token": 3e-05, + "input_cost_per_token_priority": 6e-05, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.8e-04, + "output_cost_per_token_priority": 2.7e-04, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": false, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true, + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true + }, + "gpt-5.4-pro-2026-03-05": { + "cache_read_input_token_cost": 3e-06, + "cache_read_input_token_cost_priority": 6e-06, + "input_cost_per_token": 3e-05, + "input_cost_per_token_priority": 6e-05, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.8e-04, + "output_cost_per_token_priority": 2.7e-04, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": false, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true, + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5-pro": { "input_cost_per_token": 1.5e-05, @@ -20423,7 +20633,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-pro-2025-10-06": { "input_cost_per_token": 1.5e-05, @@ -20456,7 +20668,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-2025-08-07": { "cache_read_input_token_cost": 1.25e-07, @@ -20495,7 +20709,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-chat": { "cache_read_input_token_cost": 1.25e-07, @@ -20527,7 +20743,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": false, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-chat-latest": { "cache_read_input_token_cost": 1.25e-07, @@ -20559,7 +20777,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": false, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-codex": { "cache_read_input_token_cost": 1.25e-07, @@ -20589,7 +20809,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-codex": { "cache_read_input_token_cost": 1.25e-07, @@ -20622,7 +20844,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-codex-max": { "cache_read_input_token_cost": 1.25e-07, @@ -20652,7 +20876,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5.1-codex-mini": { "cache_read_input_token_cost": 2.5e-08, @@ -20685,7 +20911,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.2-codex": { "cache_read_input_token_cost": 1.75e-07, @@ -20718,7 +20946,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5.3-codex": { "cache_read_input_token_cost": 1.75e-07, @@ -20751,7 +20981,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-mini": { "cache_read_input_token_cost": 2.5e-08, @@ -20790,7 +21022,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-mini-2025-08-07": { "cache_read_input_token_cost": 2.5e-08, @@ -20829,7 +21063,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-nano": { "cache_read_input_token_cost": 5e-09, @@ -20865,7 +21101,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-nano-2025-08-07": { "cache_read_input_token_cost": 5e-09, @@ -20900,7 +21138,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-image-1": { "cache_read_input_image_token_cost": 2.5e-06, @@ -37352,7 +37592,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-search-api-2025-10-14": { "cache_read_input_token_cost": 1.25e-07, @@ -37371,7 +37613,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-realtime-mini-2025-10-06": { "cache_creation_input_audio_token_cost": 3e-07, diff --git a/litellm/types/utils.py b/litellm/types/utils.py index 1b7089d7f10..1970ed3949e 100644 --- a/litellm/types/utils.py +++ b/litellm/types/utils.py @@ -129,6 +129,8 @@ class ProviderSpecificModelInfo(TypedDict, total=False): supports_web_search: Optional[bool] supports_reasoning: Optional[bool] supports_url_context: Optional[bool] + supports_none_reasoning_effort: Optional[bool] + supports_xhigh_reasoning_effort: Optional[bool] class SearchContextCostPerQuery(TypedDict, total=False): diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index f52288ea72a..3bacdfddf2f 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -20113,7 +20113,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1": { "cache_read_input_token_cost": 1.25e-07, @@ -20149,7 +20151,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-2025-11-13": { "cache_read_input_token_cost": 1.25e-07, @@ -20185,7 +20189,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-chat-latest": { "cache_read_input_token_cost": 1.25e-07, @@ -20220,7 +20226,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": false, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": false }, "gpt-5.2": { "cache_read_input_token_cost": 1.75e-07, @@ -20257,7 +20265,10 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": true, + "supports_web_search": true }, "gpt-5.2-2025-12-11": { "cache_read_input_token_cost": 1.75e-07, @@ -20294,7 +20305,10 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": true, + "supports_web_search": true }, "gpt-5.2-chat-latest": { "cache_read_input_token_cost": 1.75e-07, @@ -20328,7 +20342,47 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false, + "supports_web_search": true + }, + "gpt-5.3-chat-latest": { + "cache_read_input_token_cost": 1.75e-07, + "cache_read_input_token_cost_priority": 3.5e-07, + "input_cost_per_token": 1.75e-06, + "input_cost_per_token_priority": 3.5e-06, + "litellm_provider": "openai", + "max_input_tokens": 128000, + "max_output_tokens": 16384, + "max_tokens": 16384, + "mode": "chat", + "output_cost_per_token": 1.4e-05, + "output_cost_per_token_priority": 2.8e-05, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false, + "supports_web_search": true }, "gpt-5.2-pro": { "input_cost_per_token": 2.1e-05, @@ -20359,7 +20413,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5.2-pro-2025-12-11": { "input_cost_per_token": 2.1e-05, @@ -20390,7 +20446,161 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true + }, + "gpt-5.4": { + "cache_read_input_token_cost": 2.5e-07, + "cache_read_input_token_cost_priority": 5e-07, + "input_cost_per_token": 2.5e-06, + "input_cost_per_token_priority": 5e-06, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.5e-05, + "output_cost_per_token_priority": 2.25e-05, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true, + "supports_none_reasoning_effort": true, + "supports_xhigh_reasoning_effort": true + }, + "gpt-5.4-2026-03-05": { + "cache_read_input_token_cost": 2.5e-07, + "cache_read_input_token_cost_priority": 5e-07, + "input_cost_per_token": 2.5e-06, + "input_cost_per_token_priority": 5e-06, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.5e-05, + "output_cost_per_token_priority": 2.25e-05, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true + }, + "gpt-5.4-pro": { + "cache_read_input_token_cost": 3e-06, + "cache_read_input_token_cost_priority": 6e-06, + "input_cost_per_token": 3e-05, + "input_cost_per_token_priority": 6e-05, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.8e-04, + "output_cost_per_token_priority": 2.7e-04, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": false, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true, + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true + }, + "gpt-5.4-pro-2026-03-05": { + "cache_read_input_token_cost": 3e-06, + "cache_read_input_token_cost_priority": 6e-06, + "input_cost_per_token": 3e-05, + "input_cost_per_token_priority": 6e-05, + "litellm_provider": "openai", + "max_input_tokens": 1050000, + "max_output_tokens": 128000, + "max_tokens": 128000, + "mode": "responses", + "output_cost_per_token": 1.8e-04, + "output_cost_per_token_priority": 2.7e-04, + "supported_endpoints": [ + "/v1/chat/completions", + "/v1/batch", + "/v1/responses" + ], + "supported_modalities": [ + "text", + "image" + ], + "supported_output_modalities": [ + "text" + ], + "supports_function_calling": true, + "supports_native_streaming": true, + "supports_parallel_function_calling": true, + "supports_pdf_input": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_response_schema": false, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_service_tier": true, + "supports_vision": true, + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5-pro": { "input_cost_per_token": 1.5e-05, @@ -20423,7 +20633,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-pro-2025-10-06": { "input_cost_per_token": 1.5e-05, @@ -20456,7 +20668,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-2025-08-07": { "cache_read_input_token_cost": 1.25e-07, @@ -20495,7 +20709,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-chat": { "cache_read_input_token_cost": 1.25e-07, @@ -20527,7 +20743,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": false, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-chat-latest": { "cache_read_input_token_cost": 1.25e-07, @@ -20559,7 +20777,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": false, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-codex": { "cache_read_input_token_cost": 1.25e-07, @@ -20589,7 +20809,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-codex": { "cache_read_input_token_cost": 1.25e-07, @@ -20622,7 +20844,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.1-codex-max": { "cache_read_input_token_cost": 1.25e-07, @@ -20652,7 +20876,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5.1-codex-mini": { "cache_read_input_token_cost": 2.5e-08, @@ -20685,7 +20911,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5.2-codex": { "cache_read_input_token_cost": 1.75e-07, @@ -20718,7 +20946,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": true }, "gpt-5.3-codex": { "cache_read_input_token_cost": 1.75e-07, @@ -20751,7 +20981,9 @@ "supports_response_schema": true, "supports_system_messages": false, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-mini": { "cache_read_input_token_cost": 2.5e-08, @@ -20790,7 +21022,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-mini-2025-08-07": { "cache_read_input_token_cost": 2.5e-08, @@ -20829,7 +21063,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_service_tier": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-nano": { "cache_read_input_token_cost": 5e-09, @@ -20865,7 +21101,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-nano-2025-08-07": { "cache_read_input_token_cost": 5e-09, @@ -20900,7 +21138,9 @@ "supports_response_schema": true, "supports_system_messages": true, "supports_tool_choice": true, - "supports_vision": true + "supports_vision": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-image-1": { "cache_read_input_image_token_cost": 2.5e-06, @@ -37352,7 +37592,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-5-search-api-2025-10-14": { "cache_read_input_token_cost": 1.25e-07, @@ -37371,7 +37613,9 @@ "supports_system_messages": true, "supports_tool_choice": true, "supports_vision": true, - "supports_web_search": true + "supports_web_search": true, + "supports_none_reasoning_effort": false, + "supports_xhigh_reasoning_effort": false }, "gpt-realtime-mini-2025-10-06": { "cache_creation_input_audio_token_cost": 3e-07, diff --git a/tests/test_litellm/llms/openai/test_gpt5_transformation.py b/tests/test_litellm/llms/openai/test_gpt5_transformation.py index 386f264a4dd..4b64239d0e9 100644 --- a/tests/test_litellm/llms/openai/test_gpt5_transformation.py +++ b/tests/test_litellm/llms/openai/test_gpt5_transformation.py @@ -1,8 +1,8 @@ import pytest import litellm -from litellm.llms.openai.openai import OpenAIConfig from litellm.llms.openai.chat.gpt_5_transformation import OpenAIGPT5Config +from litellm.llms.openai.openai import OpenAIConfig @pytest.fixture() @@ -260,18 +260,21 @@ def test_gpt5_drops_reasoning_effort_xhigh_when_requested(config: OpenAIConfig): # GPT-5.1 temperature handling tests def test_gpt5_1_model_detection(gpt5_config: OpenAIGPT5Config): - """Test that GPT-5.1 models are correctly detected.""" - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.1") - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.1-codex") - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.1-codex-max") - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.1-chat") - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.2") - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.2-2025-12-11") - assert gpt5_config.is_model_gpt_5_1_model("gpt-5.2-chat-latest") - assert not gpt5_config.is_model_gpt_5_1_model("gpt-5.2-pro") - assert not gpt5_config.is_model_gpt_5_1_model("gpt-5") - assert not gpt5_config.is_model_gpt_5_1_model("gpt-5-mini") - assert not gpt5_config.is_model_gpt_5_1_model("gpt-5-codex") + """Test that models supporting reasoning_effort='none' are correctly detected via model map.""" + # gpt-5.1 and gpt-5.2 chat variants support none + assert gpt5_config._supports_reasoning_effort_level("gpt-5.1", "none") + assert gpt5_config._supports_reasoning_effort_level("gpt-5.1-2025-11-13", "none") + assert gpt5_config._supports_reasoning_effort_level("gpt-5.1-chat-latest", "none") + assert gpt5_config._supports_reasoning_effort_level("gpt-5.2", "none") + assert gpt5_config._supports_reasoning_effort_level("gpt-5.2-2025-12-11", "none") + # codex/pro/chat variants do not support none + assert not gpt5_config._supports_reasoning_effort_level("gpt-5.1-codex", "none") + assert not gpt5_config._supports_reasoning_effort_level("gpt-5.1-codex-max", "none") + assert not gpt5_config._supports_reasoning_effort_level("gpt-5.2-chat-latest", "none") + assert not gpt5_config._supports_reasoning_effort_level("gpt-5.2-pro", "none") + assert not gpt5_config._supports_reasoning_effort_level("gpt-5", "none") + assert not gpt5_config._supports_reasoning_effort_level("gpt-5-mini", "none") + assert not gpt5_config._supports_reasoning_effort_level("gpt-5-codex", "none") def test_gpt5_1_temperature_with_reasoning_effort_none(config: OpenAIConfig): @@ -301,6 +304,36 @@ def test_gpt5_2_temperature_with_reasoning_effort_none(config: OpenAIConfig): assert params["reasoning_effort"] == "none" +def test_gpt5_4_allows_reasoning_effort_xhigh(config: OpenAIConfig): + params = config.map_openai_params( + non_default_params={"reasoning_effort": "xhigh"}, + optional_params={}, + model="gpt-5.4", + drop_params=False, + ) + assert params["reasoning_effort"] == "xhigh" + + +def test_gpt5_4_pro_allows_reasoning_effort_xhigh(config: OpenAIConfig): + params = config.map_openai_params( + non_default_params={"reasoning_effort": "xhigh"}, + optional_params={}, + model="gpt-5.4-pro", + drop_params=False, + ) + assert params["reasoning_effort"] == "xhigh" + + +def test_gpt5_4_pro_rejects_non_default_temperature(config: OpenAIConfig): + with pytest.raises(litellm.utils.UnsupportedParamsError): + config.map_openai_params( + non_default_params={"temperature": 0.5}, + optional_params={}, + model="gpt-5.4-pro", + drop_params=False, + ) + + def test_gpt5_1_temperature_without_reasoning_effort(config: OpenAIConfig): """Test that GPT-5.1 supports any temperature when reasoning_effort is not specified. @@ -395,7 +428,38 @@ def test_gpt5_temperature_still_restricted(config: OpenAIConfig): assert params["temperature"] == 1.0 -def test_gpt5_2_pro_allows_reasoning_effort_xhigh(config: OpenAIConfig): +def test_gpt5_2_chat_temperature_restricted(config: OpenAIConfig): + """Test that gpt-5.2-chat only supports temperature=1, like base gpt-5. + + Regression test for https://github.com/BerriAI/litellm/issues/21911 + """ + # gpt-5.2-chat should reject non-1 temperature when drop_params=False + for model in ["gpt-5.2-chat", "gpt-5.2-chat-latest"]: + with pytest.raises(litellm.utils.UnsupportedParamsError): + config.map_openai_params( + non_default_params={"temperature": 0.7}, + optional_params={}, + model=model, + drop_params=False, + ) + + # temperature=1 should still work + params = config.map_openai_params( + non_default_params={"temperature": 1.0}, + optional_params={}, + model=model, + drop_params=False, + ) + assert params["temperature"] == 1.0 + + # drop_params=True should silently drop non-1 temperature + params = config.map_openai_params( + non_default_params={"temperature": 0.5}, + optional_params={}, + model=model, + drop_params=True, + ) + assert "temperature" not in params params = config.map_openai_params( non_default_params={"reasoning_effort": "xhigh"}, optional_params={}, @@ -414,3 +478,174 @@ def test_gpt5_2_allows_reasoning_effort_xhigh(config: OpenAIConfig): drop_params=False, ) assert params["reasoning_effort"] == "xhigh" + + +# GPT-5-Search specific tests +def test_gpt5_search_model_detection(gpt5_config: OpenAIGPT5Config): + """Test that GPT-5 search models are correctly detected.""" + assert gpt5_config.is_model_gpt_5_search_model("gpt-5-search-api") + assert gpt5_config.is_model_gpt_5_search_model("gpt-5-search-mini-api") + + assert not gpt5_config.is_model_gpt_5_search_model("gpt-5") + assert not gpt5_config.is_model_gpt_5_search_model("gpt-5-codex") + assert not gpt5_config.is_model_gpt_5_search_model("gpt-5-mini") + + +def test_gpt5_search_supported_params(gpt5_config: OpenAIGPT5Config): + """Test that search models do NOT list reasoning/tool params as supported.""" + supported = gpt5_config.get_supported_openai_params(model="gpt-5-search-api") + rejected = [ + "logit_bias", + "modalities", + "prediction", + "n", + "seed", + "temperature", + "tools", + "tool_choice", + "function_call", + "functions", + "parallel_tool_calls", + "audio", + "reasoning_effort", + ] + for param in rejected: + assert param not in supported, f"{param} should not be supported for search models" + + +def test_gpt5_search_has_expected_params(gpt5_config: OpenAIGPT5Config): + """Test that search models DO list the correct supported params.""" + supported = gpt5_config.get_supported_openai_params(model="gpt-5-search-api") + expected = [ + "max_tokens", + "max_completion_tokens", + "stream", + "stream_options", + "web_search_options", + "service_tier", + "response_format", + "user", + "store", + "verbosity", + "extra_headers", + ] + for param in expected: + assert param in supported, f"{param} should be supported for search models" + + +def test_gpt5_search_maps_max_tokens(config: OpenAIConfig): + """Test that search models map max_tokens -> max_completion_tokens.""" + params = config.map_openai_params( + non_default_params={"max_tokens": 200}, + optional_params={}, + model="gpt-5-search-api", + drop_params=False, + ) + assert params["max_completion_tokens"] == 200 + assert "max_tokens" not in params + + +def test_gpt5_search_drops_unsupported_params(config: OpenAIConfig): + """Test that search models drop unsupported params via map_openai_params.""" + params = config.map_openai_params( + non_default_params={"n": 2, "temperature": 0.7, "tools": [{"type": "function"}]}, + optional_params={}, + model="gpt-5-search-api", + drop_params=True, + ) + assert "n" not in params + assert "temperature" not in params + assert "tools" not in params +# GPT-5 unsupported params audit (validated via direct API calls) +def test_gpt5_rejects_params_unsupported_by_openai(config: OpenAIConfig): + """Params that OpenAI rejects for all GPT-5 reasoning models.""" + rejected_params = [ + "logit_bias", + "modalities", + "prediction", + "audio", + "web_search_options", + ] + for model in ["gpt-5", "gpt-5-mini", "gpt-5-codex", "gpt-5.1", "gpt-5.2"]: + supported = config.get_supported_openai_params(model=model) + for param in rejected_params: + assert param not in supported, ( + f"{param} should not be supported for {model}" + ) + + +def test_gpt5_1_supports_logprobs_top_p(config: OpenAIConfig): + """gpt-5.1/5.2 support logprobs, top_p, top_logprobs when reasoning_effort='none'.""" + for model in ["gpt-5.1", "gpt-5.2"]: + supported = config.get_supported_openai_params(model=model) + assert "logprobs" in supported, f"logprobs should be supported for {model}" + assert "top_p" in supported, f"top_p should be supported for {model}" + assert "top_logprobs" in supported, f"top_logprobs should be supported for {model}" + + +def test_gpt5_base_does_not_support_logprobs_top_p(config: OpenAIConfig): + """Base gpt-5/gpt-5-mini do NOT support logprobs, top_p, top_logprobs.""" + for model in ["gpt-5", "gpt-5-mini", "gpt-5-codex"]: + supported = config.get_supported_openai_params(model=model) + assert "logprobs" not in supported, f"logprobs should not be supported for {model}" + assert "top_p" not in supported, f"top_p should not be supported for {model}" + assert "top_logprobs" not in supported, f"top_logprobs should not be supported for {model}" + + +def test_gpt5_1_logprobs_passthrough(config: OpenAIConfig): + """Test that logprobs passes through for gpt-5.1.""" + params = config.map_openai_params( + non_default_params={"logprobs": True, "top_logprobs": 3}, + optional_params={}, + model="gpt-5.1", + drop_params=False, + ) + assert params["logprobs"] is True + assert params["top_logprobs"] == 3 + + +def test_gpt5_1_top_p_passthrough(config: OpenAIConfig): + """Test that top_p passes through for gpt-5.1.""" + params = config.map_openai_params( + non_default_params={"top_p": 0.9}, + optional_params={}, + model="gpt-5.1", + drop_params=False, + ) + assert params["top_p"] == 0.9 + + +def test_gpt5_1_logprobs_rejected_with_reasoning_effort(config: OpenAIConfig): + """logprobs/top_p/top_logprobs are rejected when reasoning_effort != 'none'.""" + for effort in ["low", "medium", "high"]: + with pytest.raises(litellm.utils.UnsupportedParamsError): + config.map_openai_params( + non_default_params={"logprobs": True, "reasoning_effort": effort}, + optional_params={}, + model="gpt-5.1", + drop_params=False, + ) + + +def test_gpt5_1_top_p_rejected_with_reasoning_effort(config: OpenAIConfig): + """top_p is rejected when reasoning_effort != 'none'.""" + with pytest.raises(litellm.utils.UnsupportedParamsError): + config.map_openai_params( + non_default_params={"top_p": 0.9, "reasoning_effort": "high"}, + optional_params={}, + model="gpt-5.1", + drop_params=False, + ) + + +def test_gpt5_1_logprobs_dropped_with_reasoning_effort(config: OpenAIConfig): + """logprobs/top_p are dropped when reasoning_effort != 'none' and drop_params=True.""" + params = config.map_openai_params( + non_default_params={"logprobs": True, "top_p": 0.9, "reasoning_effort": "high"}, + optional_params={}, + model="gpt-5.1", + drop_params=True, + ) + assert "logprobs" not in params + assert "top_p" not in params + assert params["reasoning_effort"] == "high" \ No newline at end of file