diff --git a/litellm/llms/azure/responses/o_series_transformation.py b/litellm/llms/azure/responses/o_series_transformation.py index 06346a300e4..38e524ac8bb 100644 --- a/litellm/llms/azure/responses/o_series_transformation.py +++ b/litellm/llms/azure/responses/o_series_transformation.py @@ -8,11 +8,12 @@ Translations handled by LiteLLM: - Other parameters follow base Azure OpenAI Responses API behavior """ +from collections.abc import Sequence from typing import TYPE_CHECKING, Any, Final from litellm._logging import verbose_logger from litellm.types.llms.openai import ResponsesAPIOptionalRequestParams -from litellm.utils import supports_reasoning +from litellm.utils import has_catalog_entry, supports_reasoning from .transformation import AzureOpenAIResponsesAPIConfig @@ -39,7 +40,9 @@ class AzureOpenAIOSeriesResponsesAPIConfig(AzureOpenAIResponsesAPIConfig): O-series models don't support temperature parameter in responses API. """ # Get the base Azure supported params - base_supported_params: Final = super().get_supported_openai_params(model) + base_supported_params: Final[Sequence[str]] = super().get_supported_openai_params(model) + if has_catalog_entry(model=model, custom_llm_provider="azure") and self._sampling_params_allowed(model, None): + return list(base_supported_params) # O-series models don't support temperature parameter in responses API o_series_unsupported_params: Final = ["temperature"] @@ -63,6 +66,13 @@ class AzureOpenAIOSeriesResponsesAPIConfig(AzureOpenAIResponsesAPIConfig): Drops temperature parameter if drop_params is True since O-series models don't support temperature in the responses API. """ + if has_catalog_entry(model=model, custom_llm_provider="azure"): + return dict( + super().map_openai_params_for_deployment( + response_api_optional_params, model, drop_params, declared_reasoning_efforts=None + ) + ) + mapped_params: Final = dict(response_api_optional_params) # If drop_params is enabled, remove temperature parameter for O-series models diff --git a/litellm/llms/azure/responses/transformation.py b/litellm/llms/azure/responses/transformation.py index a814a9fbac0..73187c456bb 100644 --- a/litellm/llms/azure/responses/transformation.py +++ b/litellm/llms/azure/responses/transformation.py @@ -97,6 +97,11 @@ class AzureOpenAIResponsesAPIConfig(OpenAIResponsesAPIConfig): def custom_llm_provider(self) -> LlmProviders: return LlmProviders.AZURE + def _sampling_params_allowed(self, model: str, effort: str | None) -> bool: + return AzureOpenAIGPT5Config.sampling_params_allowed( + model, effort, fallback_allowed=not self._is_gpt_5_model(model) + ) + @staticmethod def _is_gpt_5_model(model: str) -> bool: return AzureOpenAIGPT5Config.is_model_gpt_5_model(model) diff --git a/litellm/llms/openai/chat/gpt_reasoning_transformation.py b/litellm/llms/openai/chat/gpt_reasoning_transformation.py index 8c9551d01c0..0243d771834 100644 --- a/litellm/llms/openai/chat/gpt_reasoning_transformation.py +++ b/litellm/llms/openai/chat/gpt_reasoning_transformation.py @@ -11,6 +11,7 @@ from litellm.utils import ( has_catalog_entry, is_explicitly_disabled_factory, supports_factory, + supports_sampling_params, ) from .gpt_transformation import OpenAIGPTConfig @@ -205,6 +206,14 @@ class OpenAIGPT5Config(OpenAIGPTConfig): ) -> bool: return cls._supports_reasoning_effort_level(model, level) + @classmethod + def sampling_params_allowed(cls, model: str, effort: str | None, *, fallback_allowed: bool) -> bool: + lookup: Final = cls._model_map_lookup_name(model) + if cls._supports_reasoning_effort_level(model, "none") and cls.effort_resolves_to_none(model, effort): + return True + flag: Final = supports_sampling_params(lookup) + return fallback_allowed if flag is None else flag + @classmethod def effort_resolves_to_none(cls, model: str, effective_effort: str | None) -> bool: """Whether this request's reasoning effort ends up as "none", which is the single @@ -318,7 +327,9 @@ class OpenAIGPT5Config(OpenAIGPTConfig): # 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"]) + non_supported_params.extend(["logprobs", "top_logprobs"]) + if not self.sampling_params_allowed(model, None, fallback_allowed=False): + non_supported_params.append("top_p") return [param for param in base_gpt_series_params if param not in non_supported_params] @@ -398,7 +409,7 @@ class OpenAIGPT5Config(OpenAIGPTConfig): # gpt-5.1/5.2 support logprobs, top_p, top_logprobs only when reasoning_effort="none" supports_none: Final = self._supports_reasoning_effort_level(model, "none") if supports_none: - sampling_params: Final = ["logprobs", "top_logprobs", "top_p"] + sampling_params: Final = ["logprobs", "top_logprobs"] has_sampling: Final = any(p in non_default_params for p in sampling_params) if has_sampling and not self.effort_resolves_to_none(model, effective_effort): if litellm.drop_params or drop_params: @@ -415,12 +426,19 @@ class OpenAIGPT5Config(OpenAIGPTConfig): status_code=400, ) + sampling_allowed: Final = self.sampling_params_allowed(model, effective_effort, fallback_allowed=False) + if "top_p" in non_default_params and not sampling_allowed: + if litellm.drop_params or drop_params: + non_default_params.pop("top_p", None) + else: + raise litellm.utils.UnsupportedParamsError( + message=f"{model} only supports top_p when reasoning_effort resolves to 'none'", + status_code=400, + ) if "temperature" in non_default_params: temperature_value: Final[float | None] = non_default_params.pop("temperature") if temperature_value is not None: - # a non-default temperature rides on the effort resolving to "none", not on - # the model merely supporting it - if (supports_none and self.effort_resolves_to_none(model, effective_effort)) or temperature_value == 1: + if sampling_allowed or temperature_value == 1: optional_params["temperature"] = temperature_value elif litellm.drop_params or drop_params: pass diff --git a/litellm/llms/openai/chat/o_series_transformation.py b/litellm/llms/openai/chat/o_series_transformation.py index 4d56a3b38c5..b95e6a75ab0 100644 --- a/litellm/llms/openai/chat/o_series_transformation.py +++ b/litellm/llms/openai/chat/o_series_transformation.py @@ -25,6 +25,7 @@ from litellm.utils import ( supports_system_messages, ) +from .gpt_reasoning_transformation import OpenAIGPT5Config from .gpt_transformation import OpenAIGPTConfig @@ -60,7 +61,6 @@ class OpenAIOSeriesConfig(OpenAIGPTConfig): all_openai_params: Final = super().get_supported_openai_params(model=model) non_supported_params: Final = [ "logprobs", - "top_p", "presence_penalty", "frequency_penalty", "top_logprobs", @@ -68,6 +68,9 @@ class OpenAIOSeriesConfig(OpenAIGPTConfig): o_series_only_param: Final = ["reasoning_effort"] + if not OpenAIGPT5Config.sampling_params_allowed(model, None, fallback_allowed=False): + non_supported_params.append("top_p") + all_openai_params.extend(o_series_only_param) try: @@ -105,10 +108,22 @@ class OpenAIOSeriesConfig(OpenAIGPTConfig): ): if "max_tokens" in non_default_params: optional_params["max_completion_tokens"] = non_default_params.pop("max_tokens") + sampling_params: Final = cast( # cast-ok: legacy mapper accepts an untyped parameter bag + "dict[str, object]", non_default_params + ) + reasoning_effort: Final = sampling_params.get("reasoning_effort") + sampling_allowed: Final = OpenAIGPT5Config.sampling_params_allowed( + model, reasoning_effort if isinstance(reasoning_effort, str) else None, fallback_allowed=False + ) + if "top_p" in non_default_params and not sampling_allowed: + if litellm.drop_params or drop_params: + sampling_params.pop("top_p", None) + else: + raise litellm.utils.UnsupportedParamsError(message=f"{model} doesn't support top_p", status_code=400) if "temperature" in non_default_params: temperature_value: Final[float | None] = non_default_params.pop("temperature") if temperature_value is not None: - if temperature_value == 1: + if sampling_allowed or temperature_value == 1: optional_params["temperature"] = temperature_value else: ## UNSUPPORTED TOOL CHOICE VALUE diff --git a/litellm/llms/openai/responses/transformation.py b/litellm/llms/openai/responses/transformation.py index 314a1d22508..744e70d7df1 100644 --- a/litellm/llms/openai/responses/transformation.py +++ b/litellm/llms/openai/responses/transformation.py @@ -35,6 +35,7 @@ from litellm.litellm_core_utils.safe_json_loads import safe_json_loads from litellm.litellm_core_utils.url_utils import encode_url_path_segment, join_api_base from litellm.llms.base_llm.responses.transformation import BaseResponsesAPIConfig from litellm.llms.openai.chat.gpt_reasoning_transformation import ( + OpenAIGPT5Config, is_gpt_reasoning_series_name, refused_reasoning_effort_warning, unsupported_reasoning_effort_message, @@ -373,6 +374,13 @@ class OpenAIResponsesAPIConfig(BaseResponsesAPIConfig): def _model_map_lookup_name(model: str) -> str: return model + def _sampling_params_allowed(self, model: str, effort: str | None) -> bool: + if self.custom_llm_provider not in (LlmProviders.OPENAI, LlmProviders.AZURE): + return not self._is_gpt_5_model(model) or ( + self._supports_reasoning_effort_none(model) and self._effort_resolves_to_none(model, effort) + ) + return OpenAIGPT5Config.sampling_params_allowed(model, effort, fallback_allowed=not self._is_gpt_5_model(model)) + @staticmethod def _supports_reasoning_effort_none(model: str) -> bool: """Return True if the model supports reasoning.effort='none'.""" @@ -494,7 +502,7 @@ class OpenAIResponsesAPIConfig(BaseResponsesAPIConfig): ) -> Mapping[str, object]: """No mapping applied since inputs are in OpenAI spec already. - GPT-5 models have restrictions on temperature and top_p (only temperature=1 + Reasoning models have restrictions on temperature and top_p (only temperature=1 is accepted, and top_p is rejected, unless reasoning.effort resolves to 'none' on models that support it). Apply the same validation used by the chat completions path. @@ -553,42 +561,43 @@ class OpenAIResponsesAPIConfig(BaseResponsesAPIConfig): params["reasoning"] = remaining else: params.pop("reasoning", None) - forwarded_reasoning: Final = params.get("reasoning") - effort: Final = forwarded_reasoning.get("effort") if isinstance(forwarded_reasoning, dict) else None - supports_none: Final = self._supports_reasoning_effort_none(model=lookup_name) - effort_is_none: Final = supports_none and self._effort_resolves_to_none(lookup_name, effort) + forwarded_reasoning: Final = params.get("reasoning") + effort: Final = self._requested_effort(forwarded_reasoning) + sampling_allowed: Final = self._sampling_params_allowed( + lookup_name, effort if isinstance(effort, str) else None + ) - temperature: Final = params.get("temperature") - if temperature is not None and temperature != 1: - if effort_is_none: - pass # flexible temperature allowed - elif drop_params or litellm.drop_params: - params.pop("temperature", None) - else: - raise litellm.UnsupportedParamsError( - message=( - f"{model} doesn't support temperature={temperature} while reasoning is " - "active. Only temperature=1 is supported unless reasoning.effort resolves " - "to 'none', either set explicitly on the request or declared as the " - "model's default_reasoning_effort. " - "To drop unsupported params set `litellm.drop_params = True`" - ), - status_code=400, - ) + temperature: Final = params.get("temperature") + if temperature is not None and temperature != 1: + if sampling_allowed: + pass + elif drop_params or litellm.drop_params: + params.pop("temperature", None) + else: + raise litellm.UnsupportedParamsError( + message=( + f"{model} doesn't support temperature={temperature} while reasoning is " + "active. Only temperature=1 is supported unless reasoning.effort resolves " + "to 'none', either set explicitly on the request or declared as the " + "model's default_reasoning_effort. " + "To drop unsupported params set `litellm.drop_params = True`" + ), + status_code=400, + ) - if "top_p" in params and not effort_is_none: - if drop_params or litellm.drop_params: - params.pop("top_p", None) - else: - raise litellm.UnsupportedParamsError( - message=( - f"{model} only supports top_p when reasoning.effort resolves to 'none', " - "either set explicitly on the request or declared as the model's " - "default_reasoning_effort. " - "To drop unsupported params set `litellm.drop_params = True`" - ), - status_code=400, - ) + if "top_p" in params and not sampling_allowed: + if drop_params or litellm.drop_params: + params.pop("top_p", None) + else: + raise litellm.UnsupportedParamsError( + message=( + f"{model} only supports top_p when reasoning.effort resolves to 'none', " + "either set explicitly on the request or declared as the model's " + "default_reasoning_effort. " + "To drop unsupported params set `litellm.drop_params = True`" + ), + status_code=400, + ) return params diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 4e81afb0ea6..ea445dcdbd4 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -5320,6 +5320,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "cache_read_input_token_cost_priority": 2.75e-07, "deprecation_date": "2027-02-09", @@ -5366,6 +5367,7 @@ "cache_read_input_token_cost_batches": 6.875e-08 }, "azure/eu/gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-08, "cache_read_input_token_cost_priority": 4.95e-08, "deprecation_date": "2027-07-13", @@ -5412,6 +5414,7 @@ "cache_read_input_token_cost_batches": 1.375e-08 }, "azure/eu/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.2", @@ -5461,6 +5464,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.2-codex", @@ -5494,6 +5498,7 @@ "supports_vision": true }, "azure/eu/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.75e-08, @@ -5526,6 +5531,7 @@ "supports_vision": true }, "azure/eu/gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-09, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -5569,6 +5575,7 @@ "cache_read_input_token_cost_batches": 2.75e-09 }, "azure/eu/o1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", @@ -5588,6 +5595,7 @@ "supports_vision": true }, "azure/eu/o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", @@ -5607,6 +5615,7 @@ "supports_vision": true }, "azure/eu/o1-mini-2024-09-12": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -5623,6 +5632,7 @@ "supports_vision": false }, "azure/eu/o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -5642,6 +5652,7 @@ "supports_vision": false }, "azure/eu/o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -5762,6 +5773,7 @@ "supports_vision": true }, "azure/global/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.25e-07, @@ -5810,6 +5822,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/global/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.25e-07, @@ -5842,6 +5855,7 @@ "supports_vision": true }, "azure/global/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.5e-08, @@ -7062,6 +7076,7 @@ "supports_audio_input": true }, "azure/gpt-5.1-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_priority": 2.5e-07, "deprecation_date": "2027-07-13", @@ -7111,6 +7126,7 @@ "cache_read_input_token_cost_batches": 6.25e-08 }, "azure/gpt-5.1-codex-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_priority": 2.5e-07, "deprecation_date": "2027-05-15", @@ -7147,6 +7163,7 @@ "supports_vision": true }, "azure/gpt-5.1-codex-mini-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_priority": 4.5e-08, "deprecation_date": "2027-05-15", @@ -7182,6 +7199,7 @@ "supports_vision": true }, "azure/gpt-5": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.1", @@ -7228,6 +7246,7 @@ "supports_vision": true }, "azure/gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_priority": 2.5e-07, "deprecation_date": "2027-02-09", @@ -7274,6 +7293,7 @@ "cache_read_input_token_cost_batches": 6.25e-08 }, "azure/gpt-5-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2027-03-17", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7307,6 +7327,7 @@ "supports_vision": true }, "azure/gpt-5-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.4-mini", @@ -7353,6 +7374,7 @@ "supports_vision": true }, "azure/gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_priority": 4.5e-08, "deprecation_date": "2027-07-13", @@ -7399,6 +7421,7 @@ "cache_read_input_token_cost_batches": 1.25e-08 }, "azure/gpt-5-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.4-nano", @@ -7442,6 +7465,7 @@ "supports_vision": true }, "azure/gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-09, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7485,6 +7509,7 @@ "cache_read_input_token_cost_batches": 2.5e-09 }, "azure/gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-04-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.2-pro", @@ -7523,6 +7548,7 @@ "supports_vision": true }, "azure/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.2", @@ -7598,6 +7624,7 @@ "supports_vision": true }, "azure/gpt-5.1-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7621,6 +7648,7 @@ "supports_tool_choice": true }, "azure/gpt-5.2-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7643,6 +7671,7 @@ "supports_tool_choice": true }, "azure/gpt-5.3-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7665,6 +7694,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5.1-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7688,6 +7718,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5.2-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7710,6 +7741,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5.3-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7732,6 +7764,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5.1-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7755,6 +7788,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5.2-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7777,6 +7811,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5.3-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7799,6 +7834,7 @@ "supports_tool_choice": true }, "azure/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.2-codex", @@ -7832,6 +7868,7 @@ "supports_vision": true }, "azure/gpt-5.1-codex-max": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2027-05-18", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7864,6 +7901,7 @@ "supports_vision": true }, "azure/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.5e-08, @@ -7896,6 +7934,7 @@ "supports_vision": true }, "azure/gpt-5.2": { + "supports_sampling_params": false, "deprecation_date": "2027-06-08", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.4", @@ -7943,6 +7982,7 @@ "supports_vision": true }, "azure/gpt-5.2-2025-12-11": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_priority": 3.5e-07, "deprecation_date": "2027-06-08", @@ -7990,6 +8030,7 @@ "cache_read_input_token_cost_batches": 8.75e-08 }, "azure/gpt-5.2-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -8023,6 +8064,7 @@ "supports_vision": true }, "azure/gpt-5.3-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_priority": 3.5e-07, "deprecation_date": "2027-08-24", @@ -8063,6 +8105,7 @@ "supports_vision": true }, "azure/gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "azure", @@ -8101,6 +8144,7 @@ "supports_web_search": true }, "azure/gpt-5.2-pro-2025-12-11": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "azure", @@ -8139,6 +8183,7 @@ "supports_web_search": true }, "azure/gpt-5.4": { + "supports_sampling_params": false, "deprecation_date": "2027-09-02", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.5", @@ -8198,6 +8243,7 @@ "supports_minimal_reasoning_effort": true }, "azure/us/gpt-5.4": { + "supports_sampling_params": false, "deprecation_date": "2027-09-02", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.5", @@ -8249,6 +8295,7 @@ "supports_minimal_reasoning_effort": true }, "azure/eu/gpt-5.4": { + "supports_sampling_params": false, "deprecation_date": "2027-09-02", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.5", @@ -8300,6 +8347,7 @@ "supports_minimal_reasoning_effort": true }, "azure/gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-07, "cache_read_input_token_cost_above_272k_tokens": 5e-07, "cache_read_input_token_cost_above_272k_tokens_flex": 2.5e-07, @@ -8359,6 +8407,7 @@ "output_cost_per_token_above_272k_tokens_batches": 1.125e-05 }, "azure/us/gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_above_272k_tokens": 5.5e-07, "cache_read_input_token_cost_priority": 5.5e-07, @@ -8410,6 +8459,7 @@ "output_cost_per_token_above_272k_tokens_batches": 1.2375e-05 }, "azure/eu/gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_above_272k_tokens": 5.5e-07, "cache_read_input_token_cost_priority": 5.5e-07, @@ -8461,6 +8511,7 @@ "output_cost_per_token_above_272k_tokens_batches": 1.2375e-05 }, "azure/gpt-5.4-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-09-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.5-pro", @@ -8512,6 +8563,7 @@ "supports_web_search": true }, "azure/gpt-5.4-pro-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3e-06, "cache_read_input_token_cost_above_272k_tokens": 6e-06, "deprecation_date": "2027-09-07", @@ -8563,6 +8615,7 @@ "output_cost_per_token_above_272k_tokens_batches": 0.000135 }, "azure/gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1e-05, "cache_creation_input_token_cost_priority": 1e-05, @@ -8616,6 +8669,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1e-05, "cache_creation_input_token_cost_above_272k_tokens_flex": 5e-06, @@ -8681,6 +8735,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-sol-2026-07-09": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1e-05, "cache_creation_input_token_cost_above_272k_tokens_flex": 5e-06, @@ -8746,6 +8801,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 5e-06, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-06, @@ -8811,6 +8867,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-terra-2026-07-09": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 5e-06, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-06, @@ -8876,6 +8933,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-07, "cache_creation_input_token_cost_above_272k_tokens": 5e-07, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-07, @@ -8941,6 +8999,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-luna-2026-07-09": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-07, "cache_creation_input_token_cost_above_272k_tokens": 5e-07, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-07, @@ -9843,6 +9902,7 @@ "supports_web_search": true }, "azure/us/gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -9896,6 +9956,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -9953,6 +10014,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-06, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-05, @@ -10010,6 +10072,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-07, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-07, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-06, @@ -10252,6 +10315,7 @@ "supports_web_search": true }, "azure/eu/gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -10305,6 +10369,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -10362,6 +10427,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-06, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-05, @@ -10419,6 +10485,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-07, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-07, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-06, @@ -10476,6 +10543,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.5": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6", @@ -10538,6 +10606,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.5": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.6", @@ -10594,6 +10663,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.5": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6", @@ -10650,6 +10720,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_above_272k_tokens": 1e-06, "cache_read_input_token_cost_above_272k_tokens_batches": 5e-07, @@ -10709,6 +10780,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/gpt-5.5-2026-04-24": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6", @@ -10771,6 +10843,7 @@ "output_cost_per_token_above_272k_tokens_batches": 2.25e-05 }, "azure/us/gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "cache_read_input_token_cost_above_272k_tokens": 1.1e-06, "cache_read_input_token_cost_above_272k_tokens_batches": 5.5e-07, @@ -10824,6 +10897,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.5-2026-04-24": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.6", @@ -10880,6 +10954,7 @@ "output_cost_per_token_above_272k_tokens_batches": 2.475e-05 }, "azure/eu/gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "cache_read_input_token_cost_above_272k_tokens": 1.1e-06, "cache_read_input_token_cost_above_272k_tokens_batches": 5.5e-07, @@ -10933,6 +11008,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.5-2026-04-24": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6", @@ -10989,6 +11065,7 @@ "output_cost_per_token_above_272k_tokens_batches": 2.475e-05 }, "azure/gpt-5.5-pro": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3e-06, "cache_read_input_token_cost_above_272k_tokens": 6e-06, "input_cost_per_token": 3e-05, @@ -11032,6 +11109,7 @@ "supports_low_reasoning_effort": false }, "azure/gpt-5.5-pro-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3e-06, "cache_read_input_token_cost_above_272k_tokens": 6e-06, "input_cost_per_token": 3e-05, @@ -11071,6 +11149,7 @@ "supports_web_search": true }, "azure/gpt-5.4-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 7.5e-08, @@ -11123,6 +11202,7 @@ "supports_xhigh_reasoning_effort": true }, "azure/gpt-5.4-mini-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-08, "cache_read_input_token_cost_flex": 3.75e-08, "cache_read_input_token_cost_priority": 1.5e-07, @@ -11175,6 +11255,7 @@ "cache_read_input_token_cost_batches": 3.75e-08 }, "azure/gpt-5.4-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2e-08, @@ -11224,6 +11305,7 @@ "supports_xhigh_reasoning_effort": true }, "azure/gpt-5.4-nano-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2e-08, "cache_read_input_token_cost_flex": 1e-08, "deprecation_date": "2027-09-21", @@ -11651,6 +11733,7 @@ "supports_function_calling": true }, "azure/o1": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", "cache_read_input_token_cost": 7.5e-06, @@ -11668,6 +11751,7 @@ "supports_vision": true }, "azure/o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", @@ -11688,6 +11772,7 @@ "supports_vision": true }, "azure/o1-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "input_cost_per_token": 1.1e-06, "input_cost_per_token_batches": 5.5e-07, @@ -11705,6 +11790,7 @@ "supports_vision": false }, "azure/o1-mini-2024-09-12": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "input_cost_per_token": 1.1e-06, "input_cost_per_token_batches": 5.5e-07, @@ -11722,6 +11808,7 @@ "supports_vision": false }, "azure/o3": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", "cache_read_input_token_cost": 5e-07, @@ -11760,6 +11847,7 @@ "supports_vision": true }, "azure/o3-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", "cache_read_input_token_cost": 5e-07, @@ -11798,6 +11886,7 @@ "supports_vision": true }, "azure/o3-deep-research": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-06, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -11837,6 +11926,7 @@ "supports_web_search": true }, "azure/o3-mini": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-terra", @@ -11857,6 +11947,7 @@ "supports_vision": false }, "azure/o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -11876,6 +11967,7 @@ "supports_vision": false }, "azure/o3-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-sol", @@ -11913,6 +12005,7 @@ "supports_vision": true }, "azure/o3-pro-2025-06-10": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-sol", @@ -11950,6 +12043,7 @@ "supports_vision": true }, "azure/o4-mini": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-terra", @@ -11989,6 +12083,7 @@ "supports_vision": true }, "azure/o4-mini-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12388,6 +12483,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "cache_read_input_token_cost_priority": 2.75e-07, "deprecation_date": "2027-02-09", @@ -12434,6 +12530,7 @@ "cache_read_input_token_cost_batches": 6.875e-08 }, "azure/us/gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-08, "cache_read_input_token_cost_priority": 4.95e-08, "deprecation_date": "2027-07-13", @@ -12480,6 +12577,7 @@ "cache_read_input_token_cost_batches": 1.375e-08 }, "azure/us/gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-09, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12523,6 +12621,7 @@ "cache_read_input_token_cost_batches": 2.75e-09 }, "azure/us/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.2", @@ -12572,6 +12671,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.2-codex", @@ -12605,6 +12705,7 @@ "supports_vision": true }, "azure/us/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.75e-08, @@ -12637,6 +12738,7 @@ "supports_vision": true }, "azure/us/o1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", @@ -12656,6 +12758,7 @@ "supports_vision": true }, "azure/us/o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", @@ -12675,6 +12778,7 @@ "supports_vision": true }, "azure/us/o1-mini-2024-09-12": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -12691,6 +12795,7 @@ "supports_vision": false }, "azure/us/o3": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -12729,6 +12834,7 @@ "supports_vision": true }, "azure/us/o3-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -12767,6 +12873,7 @@ "supports_vision": true }, "azure/us/o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12786,6 +12893,7 @@ "supports_vision": false }, "azure/us/o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12805,6 +12913,7 @@ "supports_vision": false }, "azure/us/o4-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3.03e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12832,6 +12941,7 @@ "supports_vision": true }, "azure/us/o4-mini-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3.03e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -36905,6 +37015,7 @@ "successor_model": "gpt-image-2.5-sunburst" }, "gpt-5": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -36966,6 +37077,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37000,6 +37112,7 @@ "supports_web_search": true }, "gpt-5.1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -37062,6 +37175,7 @@ "successor_model": "gpt-6-sol" }, "gpt-5.1-codex-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37095,6 +37209,7 @@ "supports_vision": true }, "gpt-5.1-codex-max": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37129,6 +37244,7 @@ "supports_web_search": true }, "gpt-5.1-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37163,6 +37279,7 @@ "supports_web_search": true }, "gpt-5.1-chat-latest": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37197,6 +37314,7 @@ "supports_vision": true }, "gpt-5.1-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -37259,6 +37377,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.2": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_batches": 8.75e-08, "cache_read_input_token_cost_flex": 8.75e-08, @@ -37320,6 +37439,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.2-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37354,6 +37474,7 @@ "supports_web_search": true }, "gpt-5.2-chat-latest": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-08-10", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37388,6 +37509,7 @@ "supports_vision": true }, "gpt-5.2-2025-12-11": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_batches": 8.75e-08, "cache_read_input_token_cost_flex": 8.75e-08, @@ -37449,6 +37571,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "openai", @@ -37498,6 +37621,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5.2-pro-2025-12-11": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "openai", @@ -37848,6 +37972,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_12hr": 1e-05, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 2e-05, @@ -37937,6 +38062,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_12hr": 1e-05, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 2e-05, @@ -38030,6 +38156,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_12hr": 5e-06, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 1e-05, @@ -38122,6 +38249,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-07, "cache_creation_input_token_cost_above_12hr": 5e-07, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 1e-06, @@ -38214,6 +38342,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-cyber": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 1.5625e-05, "cache_creation_input_token_cost_above_12hr": 3.125e-05, "cache_creation_input_token_cost_above_272k_tokens": 3.125e-05, @@ -38452,6 +38581,7 @@ "supports_web_search": true }, "gpt-5.5": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_above_272k_tokens": 1e-06, "cache_read_input_token_cost_batches": 2.5e-07, @@ -38523,6 +38653,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_above_272k_tokens": 1e-06, "cache_read_input_token_cost_batches": 2.5e-07, @@ -38594,6 +38725,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.5-pro": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38654,6 +38786,7 @@ "supports_low_reasoning_effort": false }, "gpt-5.5-pro-2026-04-23": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38714,6 +38847,7 @@ "supports_low_reasoning_effort": false }, "gpt-5.4": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-07, "cache_read_input_token_cost_above_272k_tokens": 5e-07, "cache_read_input_token_cost_batches": 1.3e-07, @@ -38780,6 +38914,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-07, "cache_read_input_token_cost_above_272k_tokens": 5e-07, "cache_read_input_token_cost_batches": 1.3e-07, @@ -38846,6 +38981,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-pro": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38907,6 +39043,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-pro-2026-03-05": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38968,6 +39105,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-08, "cache_read_input_token_cost_batches": 3.75e-08, "cache_read_input_token_cost_flex": 3.75e-08, @@ -39031,6 +39169,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.4-mini-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-08, "cache_read_input_token_cost_batches": 3.75e-08, "cache_read_input_token_cost_flex": 3.75e-08, @@ -39094,6 +39233,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.4-nano": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2e-08, "cache_read_input_token_cost_batches": 1e-08, "cache_read_input_token_cost_flex": 1e-08, @@ -39156,6 +39296,7 @@ "successor_model": "gpt-6-luna" }, "gpt-5.4-nano-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2e-08, "cache_read_input_token_cost_batches": 1e-08, "cache_read_input_token_cost_flex": 1e-08, @@ -39218,6 +39359,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -39270,6 +39412,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-pro-2025-10-06": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -39322,6 +39465,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -39455,6 +39599,7 @@ "supports_vision": true }, "gpt-5.3-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_priority": 3.5e-07, "input_cost_per_token": 1.75e-06, @@ -39509,6 +39654,7 @@ "successor_model": "gpt-6-sol" }, "gpt-5.3-chat-latest": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-08-10", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -39543,6 +39689,7 @@ "supports_vision": true }, "gpt-5-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_batches": 1.25e-08, "cache_read_input_token_cost_flex": 1.25e-08, @@ -39604,6 +39751,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_batches": 1.25e-08, "cache_read_input_token_cost_flex": 1.25e-08, @@ -39665,6 +39813,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-nano": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-09, "cache_read_input_token_cost_batches": 2.5e-09, "cache_read_input_token_cost_flex": 2.5e-09, @@ -39723,6 +39872,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-09, "cache_read_input_token_cost_batches": 2.5e-09, "cache_read_input_token_cost_flex": 2.5e-09, @@ -45679,6 +45829,7 @@ "supports_vision": false }, "o1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-06, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -45704,6 +45855,7 @@ "supports_vision": true }, "o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-06, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -45729,6 +45881,7 @@ "supports_vision": true }, "o1-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -45765,6 +45918,7 @@ "supports_vision": true }, "o1-pro-2025-03-19": { + "supports_sampling_params": false, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -45801,6 +45955,7 @@ "supports_vision": true }, "o3": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_flex": 2.5e-07, "cache_read_input_token_cost_priority": 8.75e-07, @@ -45856,6 +46011,7 @@ "supports_web_search": true }, "o3-deep-research": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-06, "input_cost_per_token": 1e-05, "litellm_provider": "openai", @@ -45890,6 +46046,7 @@ "successor_model": "gpt-5.6-sol" }, "o3-deep-research-2025-06-26": { + "supports_sampling_params": false, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -45924,6 +46081,7 @@ "supports_pdf_input": true }, "o3-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_flex": 2.5e-07, "cache_read_input_token_cost_priority": 8.75e-07, @@ -45980,6 +46138,7 @@ "supports_web_search": true }, "o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -46003,6 +46162,7 @@ "supports_vision": false }, "o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -46026,6 +46186,7 @@ "supports_vision": false }, "o3-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -46073,6 +46234,7 @@ "supports_web_search": true }, "o3-pro-2025-06-10": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -46120,6 +46282,7 @@ "supports_web_search": true }, "o4-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_flex": 1.38e-07, "cache_read_input_token_cost_priority": 5e-07, @@ -46156,6 +46319,7 @@ "supports_web_search": true }, "o4-mini-deep-research": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "input_cost_per_token": 2e-06, "litellm_provider": "openai", @@ -46190,6 +46354,7 @@ "successor_model": "gpt-5.6-sol" }, "o4-mini-deep-research-2025-06-26": { + "supports_sampling_params": false, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -46224,6 +46389,7 @@ "supports_pdf_input": true }, "o4-mini-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_flex": 1.38e-07, "cache_read_input_token_cost_priority": 5e-07, @@ -59821,6 +59987,7 @@ ] }, "gpt-5-search-api": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "input_cost_per_token": 1.25e-06, "litellm_provider": "openai", @@ -59849,6 +60016,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-search-api-2025-10-14": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "input_cost_per_token": 1.25e-06, "litellm_provider": "openai", @@ -70570,6 +70738,7 @@ "source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json" }, "azure/us-gov/gpt-5.1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.71875e-07, "default_reasoning_effort": "none", "input_cost_per_token": 1.71875e-06, @@ -70604,6 +70773,7 @@ "supports_vision": true }, "azure/us-gov/o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.57e-07, "input_cost_per_token": 1.513e-06, "litellm_provider": "azure", @@ -71257,6 +71427,7 @@ "source": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing" }, "gpt-5.5-cyber": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-06, "input_cost_per_token": 1.25e-05, "litellm_provider": "openai", @@ -71267,6 +71438,7 @@ "supports_reasoning": true }, "gpt-5.5-cyber-preview": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-06, "deprecation_date": "2027-02-01", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -71725,6 +71897,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.1", @@ -71747,6 +71920,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-03-17", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.1-codex", @@ -71758,6 +71932,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.4-mini", @@ -71780,6 +71955,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.4-nano", @@ -71799,6 +71975,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-04-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.2-pro", @@ -71816,6 +71993,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.1-codex-max": { + "supports_sampling_params": false, "deprecation_date": "2027-05-18", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.375e-07, @@ -71826,6 +72004,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.2": { + "supports_sampling_params": false, "deprecation_date": "2027-06-08", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.4", @@ -71848,6 +72027,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.2-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.3-codex", @@ -71859,6 +72039,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.31e-05, "input_cost_per_token_batches": 1.155e-05, "litellm_provider": "azure", @@ -71873,6 +72054,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.3-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-08-24", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.925e-07, @@ -71891,6 +72073,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.4-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 8.25e-08, @@ -71912,6 +72095,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.4-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.2e-08, @@ -71930,6 +72114,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.4-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-09-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "input_cost_per_token": 3.3e-05, @@ -72094,6 +72279,7 @@ "supports_xhigh_reasoning_effort": true }, "azure/eu/o1-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -72104,6 +72290,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o3": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -72123,6 +72310,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o3-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -72142,6 +72330,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o3-deep-research": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6-sol", @@ -72153,6 +72342,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o4-mini": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6-terra", @@ -72173,6 +72363,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o4-mini-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6-terra", @@ -72349,6 +72540,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.1", @@ -72371,6 +72563,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-03-17", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.1-codex", @@ -72382,6 +72575,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.4-mini", @@ -72404,6 +72598,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.4-nano", @@ -72423,6 +72618,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-04-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.2-pro", @@ -72440,6 +72636,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.1-codex-max": { + "supports_sampling_params": false, "deprecation_date": "2027-05-18", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.375e-07, @@ -72450,6 +72647,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.2": { + "supports_sampling_params": false, "deprecation_date": "2027-06-08", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.4", @@ -72472,6 +72670,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.2-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.3-codex", @@ -72483,6 +72682,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.31e-05, "input_cost_per_token_batches": 1.155e-05, "litellm_provider": "azure", @@ -72497,6 +72697,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.3-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-08-24", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.925e-07, @@ -72515,6 +72716,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.4-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 8.25e-08, @@ -72536,6 +72738,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.4-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.2e-08, @@ -72554,6 +72757,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.4-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-09-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "input_cost_per_token": 3.3e-05, @@ -72574,6 +72778,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/o1-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -72584,6 +72789,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/o3-deep-research": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.6-sol", diff --git a/litellm/utils.py b/litellm/utils.py index 8944ecb5648..dec9b3f75f7 100644 --- a/litellm/utils.py +++ b/litellm/utils.py @@ -3562,6 +3562,11 @@ def supports_audio_input(model: str, custom_llm_provider: str | None = None) -> return supports_factory(model=model, custom_llm_provider=custom_llm_provider, key="supports_audio_input") +def supports_sampling_params(model: str, custom_llm_provider: str | None = None) -> bool | None: + info: Final = lookup_catalog_model_info(model, custom_llm_provider) + return None if isinstance(info, ModelInfoLookupMiss) else info.get("supports_sampling_params") + + def supports_video_input(model: str, custom_llm_provider: str | None = None) -> bool: return supports_factory(model=model, custom_llm_provider=custom_llm_provider, key="supports_video_input") @@ -7580,6 +7585,7 @@ def get_model_info_helper( supports_prompt_caching=_model_info.get("supports_prompt_caching", None), supports_prompt_cache_breakpoint=_model_info.get("supports_prompt_cache_breakpoint", None), supports_thinking_cache_preservation=_model_info.get("supports_thinking_cache_preservation", None), + supports_sampling_params=_model_info.get("supports_sampling_params", None), supports_audio_input=_model_info.get("supports_audio_input", None), supports_video_input=_model_info.get("supports_video_input", None), supports_box_questions=_model_info.get("supports_box_questions", None), diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 4e81afb0ea6..ea445dcdbd4 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -5320,6 +5320,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "cache_read_input_token_cost_priority": 2.75e-07, "deprecation_date": "2027-02-09", @@ -5366,6 +5367,7 @@ "cache_read_input_token_cost_batches": 6.875e-08 }, "azure/eu/gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-08, "cache_read_input_token_cost_priority": 4.95e-08, "deprecation_date": "2027-07-13", @@ -5412,6 +5414,7 @@ "cache_read_input_token_cost_batches": 1.375e-08 }, "azure/eu/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.2", @@ -5461,6 +5464,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.2-codex", @@ -5494,6 +5498,7 @@ "supports_vision": true }, "azure/eu/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.75e-08, @@ -5526,6 +5531,7 @@ "supports_vision": true }, "azure/eu/gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-09, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -5569,6 +5575,7 @@ "cache_read_input_token_cost_batches": 2.75e-09 }, "azure/eu/o1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", @@ -5588,6 +5595,7 @@ "supports_vision": true }, "azure/eu/o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", @@ -5607,6 +5615,7 @@ "supports_vision": true }, "azure/eu/o1-mini-2024-09-12": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -5623,6 +5632,7 @@ "supports_vision": false }, "azure/eu/o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -5642,6 +5652,7 @@ "supports_vision": false }, "azure/eu/o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -5762,6 +5773,7 @@ "supports_vision": true }, "azure/global/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.25e-07, @@ -5810,6 +5822,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/global/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.25e-07, @@ -5842,6 +5855,7 @@ "supports_vision": true }, "azure/global/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.5e-08, @@ -7062,6 +7076,7 @@ "supports_audio_input": true }, "azure/gpt-5.1-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_priority": 2.5e-07, "deprecation_date": "2027-07-13", @@ -7111,6 +7126,7 @@ "cache_read_input_token_cost_batches": 6.25e-08 }, "azure/gpt-5.1-codex-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_priority": 2.5e-07, "deprecation_date": "2027-05-15", @@ -7147,6 +7163,7 @@ "supports_vision": true }, "azure/gpt-5.1-codex-mini-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_priority": 4.5e-08, "deprecation_date": "2027-05-15", @@ -7182,6 +7199,7 @@ "supports_vision": true }, "azure/gpt-5": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.1", @@ -7228,6 +7246,7 @@ "supports_vision": true }, "azure/gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_priority": 2.5e-07, "deprecation_date": "2027-02-09", @@ -7274,6 +7293,7 @@ "cache_read_input_token_cost_batches": 6.25e-08 }, "azure/gpt-5-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2027-03-17", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7307,6 +7327,7 @@ "supports_vision": true }, "azure/gpt-5-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.4-mini", @@ -7353,6 +7374,7 @@ "supports_vision": true }, "azure/gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_priority": 4.5e-08, "deprecation_date": "2027-07-13", @@ -7399,6 +7421,7 @@ "cache_read_input_token_cost_batches": 1.25e-08 }, "azure/gpt-5-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.4-nano", @@ -7442,6 +7465,7 @@ "supports_vision": true }, "azure/gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-09, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7485,6 +7509,7 @@ "cache_read_input_token_cost_batches": 2.5e-09 }, "azure/gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-04-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.2-pro", @@ -7523,6 +7548,7 @@ "supports_vision": true }, "azure/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.2", @@ -7598,6 +7624,7 @@ "supports_vision": true }, "azure/gpt-5.1-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7621,6 +7648,7 @@ "supports_tool_choice": true }, "azure/gpt-5.2-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7643,6 +7671,7 @@ "supports_tool_choice": true }, "azure/gpt-5.3-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7665,6 +7694,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5.1-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7688,6 +7718,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5.2-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7710,6 +7741,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5.3-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7732,6 +7764,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5.1-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7755,6 +7788,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5.2-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7777,6 +7811,7 @@ "supports_tool_choice": true }, "azure/eu/gpt-5.3-chat": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.925e-07, "deprecation_date": "2026-06-29", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7799,6 +7834,7 @@ "supports_tool_choice": true }, "azure/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.2-codex", @@ -7832,6 +7868,7 @@ "supports_vision": true }, "azure/gpt-5.1-codex-max": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2027-05-18", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -7864,6 +7901,7 @@ "supports_vision": true }, "azure/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.5e-08, @@ -7896,6 +7934,7 @@ "supports_vision": true }, "azure/gpt-5.2": { + "supports_sampling_params": false, "deprecation_date": "2027-06-08", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.4", @@ -7943,6 +7982,7 @@ "supports_vision": true }, "azure/gpt-5.2-2025-12-11": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_priority": 3.5e-07, "deprecation_date": "2027-06-08", @@ -7990,6 +8030,7 @@ "cache_read_input_token_cost_batches": 8.75e-08 }, "azure/gpt-5.2-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -8023,6 +8064,7 @@ "supports_vision": true }, "azure/gpt-5.3-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_priority": 3.5e-07, "deprecation_date": "2027-08-24", @@ -8063,6 +8105,7 @@ "supports_vision": true }, "azure/gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "azure", @@ -8101,6 +8144,7 @@ "supports_web_search": true }, "azure/gpt-5.2-pro-2025-12-11": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "azure", @@ -8139,6 +8183,7 @@ "supports_web_search": true }, "azure/gpt-5.4": { + "supports_sampling_params": false, "deprecation_date": "2027-09-02", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.5", @@ -8198,6 +8243,7 @@ "supports_minimal_reasoning_effort": true }, "azure/us/gpt-5.4": { + "supports_sampling_params": false, "deprecation_date": "2027-09-02", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.5", @@ -8249,6 +8295,7 @@ "supports_minimal_reasoning_effort": true }, "azure/eu/gpt-5.4": { + "supports_sampling_params": false, "deprecation_date": "2027-09-02", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.5", @@ -8300,6 +8347,7 @@ "supports_minimal_reasoning_effort": true }, "azure/gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-07, "cache_read_input_token_cost_above_272k_tokens": 5e-07, "cache_read_input_token_cost_above_272k_tokens_flex": 2.5e-07, @@ -8359,6 +8407,7 @@ "output_cost_per_token_above_272k_tokens_batches": 1.125e-05 }, "azure/us/gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_above_272k_tokens": 5.5e-07, "cache_read_input_token_cost_priority": 5.5e-07, @@ -8410,6 +8459,7 @@ "output_cost_per_token_above_272k_tokens_batches": 1.2375e-05 }, "azure/eu/gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_above_272k_tokens": 5.5e-07, "cache_read_input_token_cost_priority": 5.5e-07, @@ -8461,6 +8511,7 @@ "output_cost_per_token_above_272k_tokens_batches": 1.2375e-05 }, "azure/gpt-5.4-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-09-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.5-pro", @@ -8512,6 +8563,7 @@ "supports_web_search": true }, "azure/gpt-5.4-pro-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3e-06, "cache_read_input_token_cost_above_272k_tokens": 6e-06, "deprecation_date": "2027-09-07", @@ -8563,6 +8615,7 @@ "output_cost_per_token_above_272k_tokens_batches": 0.000135 }, "azure/gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1e-05, "cache_creation_input_token_cost_priority": 1e-05, @@ -8616,6 +8669,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1e-05, "cache_creation_input_token_cost_above_272k_tokens_flex": 5e-06, @@ -8681,6 +8735,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-sol-2026-07-09": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1e-05, "cache_creation_input_token_cost_above_272k_tokens_flex": 5e-06, @@ -8746,6 +8801,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 5e-06, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-06, @@ -8811,6 +8867,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-terra-2026-07-09": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 5e-06, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-06, @@ -8876,6 +8933,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-07, "cache_creation_input_token_cost_above_272k_tokens": 5e-07, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-07, @@ -8941,6 +8999,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.6-luna-2026-07-09": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-07, "cache_creation_input_token_cost_above_272k_tokens": 5e-07, "cache_creation_input_token_cost_above_272k_tokens_flex": 2.5e-07, @@ -9843,6 +9902,7 @@ "supports_web_search": true }, "azure/us/gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -9896,6 +9956,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -9953,6 +10014,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-06, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-05, @@ -10010,6 +10072,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-07, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-07, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-06, @@ -10252,6 +10315,7 @@ "supports_web_search": true }, "azure/eu/gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -10305,6 +10369,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05, "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05, @@ -10362,6 +10427,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-06, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-06, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-05, @@ -10419,6 +10485,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.75e-07, "cache_creation_input_token_cost_above_272k_tokens": 5.5e-07, "cache_creation_input_token_cost_above_272k_tokens_priority": 1.1e-06, @@ -10476,6 +10543,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.5": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6", @@ -10538,6 +10606,7 @@ "supports_minimal_reasoning_effort": false }, "azure/us/gpt-5.5": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.6", @@ -10594,6 +10663,7 @@ "supports_minimal_reasoning_effort": false }, "azure/eu/gpt-5.5": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6", @@ -10650,6 +10720,7 @@ "supports_minimal_reasoning_effort": false }, "azure/gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_above_272k_tokens": 1e-06, "cache_read_input_token_cost_above_272k_tokens_batches": 5e-07, @@ -10709,6 +10780,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/gpt-5.5-2026-04-24": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6", @@ -10771,6 +10843,7 @@ "output_cost_per_token_above_272k_tokens_batches": 2.25e-05 }, "azure/us/gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "cache_read_input_token_cost_above_272k_tokens": 1.1e-06, "cache_read_input_token_cost_above_272k_tokens_batches": 5.5e-07, @@ -10824,6 +10897,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.5-2026-04-24": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.6", @@ -10880,6 +10954,7 @@ "output_cost_per_token_above_272k_tokens_batches": 2.475e-05 }, "azure/eu/gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "cache_read_input_token_cost_above_272k_tokens": 1.1e-06, "cache_read_input_token_cost_above_272k_tokens_batches": 5.5e-07, @@ -10933,6 +11008,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.5-2026-04-24": { + "supports_sampling_params": false, "deprecation_date": "2027-10-26", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6", @@ -10989,6 +11065,7 @@ "output_cost_per_token_above_272k_tokens_batches": 2.475e-05 }, "azure/gpt-5.5-pro": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3e-06, "cache_read_input_token_cost_above_272k_tokens": 6e-06, "input_cost_per_token": 3e-05, @@ -11032,6 +11109,7 @@ "supports_low_reasoning_effort": false }, "azure/gpt-5.5-pro-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3e-06, "cache_read_input_token_cost_above_272k_tokens": 6e-06, "input_cost_per_token": 3e-05, @@ -11071,6 +11149,7 @@ "supports_web_search": true }, "azure/gpt-5.4-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 7.5e-08, @@ -11123,6 +11202,7 @@ "supports_xhigh_reasoning_effort": true }, "azure/gpt-5.4-mini-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-08, "cache_read_input_token_cost_flex": 3.75e-08, "cache_read_input_token_cost_priority": 1.5e-07, @@ -11175,6 +11255,7 @@ "cache_read_input_token_cost_batches": 3.75e-08 }, "azure/gpt-5.4-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2e-08, @@ -11224,6 +11305,7 @@ "supports_xhigh_reasoning_effort": true }, "azure/gpt-5.4-nano-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2e-08, "cache_read_input_token_cost_flex": 1e-08, "deprecation_date": "2027-09-21", @@ -11651,6 +11733,7 @@ "supports_function_calling": true }, "azure/o1": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", "cache_read_input_token_cost": 7.5e-06, @@ -11668,6 +11751,7 @@ "supports_vision": true }, "azure/o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", @@ -11688,6 +11772,7 @@ "supports_vision": true }, "azure/o1-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "input_cost_per_token": 1.1e-06, "input_cost_per_token_batches": 5.5e-07, @@ -11705,6 +11790,7 @@ "supports_vision": false }, "azure/o1-mini-2024-09-12": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "input_cost_per_token": 1.1e-06, "input_cost_per_token_batches": 5.5e-07, @@ -11722,6 +11808,7 @@ "supports_vision": false }, "azure/o3": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", "cache_read_input_token_cost": 5e-07, @@ -11760,6 +11847,7 @@ "supports_vision": true }, "azure/o3-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/gpt-5.6-sol", "cache_read_input_token_cost": 5e-07, @@ -11798,6 +11886,7 @@ "supports_vision": true }, "azure/o3-deep-research": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-06, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -11837,6 +11926,7 @@ "supports_web_search": true }, "azure/o3-mini": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-terra", @@ -11857,6 +11947,7 @@ "supports_vision": false }, "azure/o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -11876,6 +11967,7 @@ "supports_vision": false }, "azure/o3-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-sol", @@ -11913,6 +12005,7 @@ "supports_vision": true }, "azure/o3-pro-2025-06-10": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-sol", @@ -11950,6 +12043,7 @@ "supports_vision": true }, "azure/o4-mini": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/gpt-5.6-terra", @@ -11989,6 +12083,7 @@ "supports_vision": true }, "azure/o4-mini-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12388,6 +12483,7 @@ "supports_tool_choice": true }, "azure/us/gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.375e-07, "cache_read_input_token_cost_priority": 2.75e-07, "deprecation_date": "2027-02-09", @@ -12434,6 +12530,7 @@ "cache_read_input_token_cost_batches": 6.875e-08 }, "azure/us/gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-08, "cache_read_input_token_cost_priority": 4.95e-08, "deprecation_date": "2027-07-13", @@ -12480,6 +12577,7 @@ "cache_read_input_token_cost_batches": 1.375e-08 }, "azure/us/gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-09, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12523,6 +12621,7 @@ "cache_read_input_token_cost_batches": 2.75e-09 }, "azure/us/gpt-5.1": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.2", @@ -12572,6 +12671,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.1-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.2-codex", @@ -12605,6 +12705,7 @@ "supports_vision": true }, "azure/us/gpt-5.1-codex-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-05-15", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.75e-08, @@ -12637,6 +12738,7 @@ "supports_vision": true }, "azure/us/o1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", @@ -12656,6 +12758,7 @@ "supports_vision": true }, "azure/us/o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 8.25e-06, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", @@ -12675,6 +12778,7 @@ "supports_vision": true }, "azure/us/o1-mini-2024-09-12": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -12691,6 +12795,7 @@ "supports_vision": false }, "azure/us/o3": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -12729,6 +12834,7 @@ "supports_vision": true }, "azure/us/o3-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/us/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -12767,6 +12873,7 @@ "supports_vision": true }, "azure/us/o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12786,6 +12893,7 @@ "supports_vision": false }, "azure/us/o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12805,6 +12913,7 @@ "supports_vision": false }, "azure/us/o4-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3.03e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -12832,6 +12941,7 @@ "supports_vision": true }, "azure/us/o4-mini-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 3.03e-07, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", @@ -36905,6 +37015,7 @@ "successor_model": "gpt-image-2.5-sunburst" }, "gpt-5": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -36966,6 +37077,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37000,6 +37112,7 @@ "supports_web_search": true }, "gpt-5.1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -37062,6 +37175,7 @@ "successor_model": "gpt-6-sol" }, "gpt-5.1-codex-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37095,6 +37209,7 @@ "supports_vision": true }, "gpt-5.1-codex-max": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37129,6 +37244,7 @@ "supports_web_search": true }, "gpt-5.1-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37163,6 +37279,7 @@ "supports_web_search": true }, "gpt-5.1-chat-latest": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37197,6 +37314,7 @@ "supports_vision": true }, "gpt-5.1-2025-11-13": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -37259,6 +37377,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.2": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_batches": 8.75e-08, "cache_read_input_token_cost_flex": 8.75e-08, @@ -37320,6 +37439,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.2-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37354,6 +37474,7 @@ "supports_web_search": true }, "gpt-5.2-chat-latest": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-08-10", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -37388,6 +37509,7 @@ "supports_vision": true }, "gpt-5.2-2025-12-11": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_batches": 8.75e-08, "cache_read_input_token_cost_flex": 8.75e-08, @@ -37449,6 +37571,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "openai", @@ -37498,6 +37621,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5.2-pro-2025-12-11": { + "supports_sampling_params": false, "input_cost_per_token": 2.1e-05, "input_cost_per_token_batches": 1.05e-05, "litellm_provider": "openai", @@ -37848,6 +37972,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_12hr": 1e-05, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 2e-05, @@ -37937,6 +38062,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-sol": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_12hr": 1e-05, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 2e-05, @@ -38030,6 +38156,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-terra": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_12hr": 5e-06, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 1e-05, @@ -38122,6 +38249,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-luna": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 2.5e-07, "cache_creation_input_token_cost_above_12hr": 5e-07, "cache_creation_input_token_cost_above_12hr_above_272k_tokens": 1e-06, @@ -38214,6 +38342,7 @@ "supports_xhigh_reasoning_effort": true }, "gpt-5.6-cyber": { + "supports_sampling_params": false, "cache_creation_input_token_cost": 1.5625e-05, "cache_creation_input_token_cost_above_12hr": 3.125e-05, "cache_creation_input_token_cost_above_272k_tokens": 3.125e-05, @@ -38452,6 +38581,7 @@ "supports_web_search": true }, "gpt-5.5": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_above_272k_tokens": 1e-06, "cache_read_input_token_cost_batches": 2.5e-07, @@ -38523,6 +38653,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.5-2026-04-23": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_above_272k_tokens": 1e-06, "cache_read_input_token_cost_batches": 2.5e-07, @@ -38594,6 +38725,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.5-pro": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38654,6 +38786,7 @@ "supports_low_reasoning_effort": false }, "gpt-5.5-pro-2026-04-23": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38714,6 +38847,7 @@ "supports_low_reasoning_effort": false }, "gpt-5.4": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-07, "cache_read_input_token_cost_above_272k_tokens": 5e-07, "cache_read_input_token_cost_batches": 1.3e-07, @@ -38780,6 +38914,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-2026-03-05": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-07, "cache_read_input_token_cost_above_272k_tokens": 5e-07, "cache_read_input_token_cost_batches": 1.3e-07, @@ -38846,6 +38981,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-pro": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38907,6 +39043,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-pro-2026-03-05": { + "supports_sampling_params": false, "input_cost_per_token": 3e-05, "input_cost_per_token_above_272k_tokens": 6e-05, "input_cost_per_token_flex": 1.5e-05, @@ -38968,6 +39105,7 @@ "source": "https://developers.openai.com/api/docs/pricing" }, "gpt-5.4-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-08, "cache_read_input_token_cost_batches": 3.75e-08, "cache_read_input_token_cost_flex": 3.75e-08, @@ -39031,6 +39169,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.4-mini-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-08, "cache_read_input_token_cost_batches": 3.75e-08, "cache_read_input_token_cost_flex": 3.75e-08, @@ -39094,6 +39233,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5.4-nano": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2e-08, "cache_read_input_token_cost_batches": 1e-08, "cache_read_input_token_cost_flex": 1e-08, @@ -39156,6 +39296,7 @@ "successor_model": "gpt-6-luna" }, "gpt-5.4-nano-2026-03-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2e-08, "cache_read_input_token_cost_batches": 1e-08, "cache_read_input_token_cost_flex": 1e-08, @@ -39218,6 +39359,7 @@ "supports_minimal_reasoning_effort": false }, "gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -39270,6 +39412,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-pro-2025-10-06": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -39322,6 +39465,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "cache_read_input_token_cost_batches": 6.25e-08, "cache_read_input_token_cost_flex": 6.25e-08, @@ -39455,6 +39599,7 @@ "supports_vision": true }, "gpt-5.3-codex": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "cache_read_input_token_cost_priority": 3.5e-07, "input_cost_per_token": 1.75e-06, @@ -39509,6 +39654,7 @@ "successor_model": "gpt-6-sol" }, "gpt-5.3-chat-latest": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.75e-07, "deprecation_date": "2026-08-10", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -39543,6 +39689,7 @@ "supports_vision": true }, "gpt-5-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_batches": 1.25e-08, "cache_read_input_token_cost_flex": 1.25e-08, @@ -39604,6 +39751,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-mini-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-08, "cache_read_input_token_cost_batches": 1.25e-08, "cache_read_input_token_cost_flex": 1.25e-08, @@ -39665,6 +39813,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-nano": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-09, "cache_read_input_token_cost_batches": 2.5e-09, "cache_read_input_token_cost_flex": 2.5e-09, @@ -39723,6 +39872,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-nano-2025-08-07": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-09, "cache_read_input_token_cost_batches": 2.5e-09, "cache_read_input_token_cost_flex": 2.5e-09, @@ -45679,6 +45829,7 @@ "supports_vision": false }, "o1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-06, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -45704,6 +45855,7 @@ "supports_vision": true }, "o1-2024-12-17": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.5e-06, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -45729,6 +45881,7 @@ "supports_vision": true }, "o1-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -45765,6 +45918,7 @@ "supports_vision": true }, "o1-pro-2025-03-19": { + "supports_sampling_params": false, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -45801,6 +45955,7 @@ "supports_vision": true }, "o3": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_flex": 2.5e-07, "cache_read_input_token_cost_priority": 8.75e-07, @@ -45856,6 +46011,7 @@ "supports_web_search": true }, "o3-deep-research": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.5e-06, "input_cost_per_token": 1e-05, "litellm_provider": "openai", @@ -45890,6 +46046,7 @@ "successor_model": "gpt-5.6-sol" }, "o3-deep-research-2025-06-26": { + "supports_sampling_params": false, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -45924,6 +46081,7 @@ "supports_pdf_input": true }, "o3-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "cache_read_input_token_cost_flex": 2.5e-07, "cache_read_input_token_cost_priority": 8.75e-07, @@ -45980,6 +46138,7 @@ "supports_web_search": true }, "o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -46003,6 +46162,7 @@ "supports_vision": false }, "o3-mini-2025-01-31": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5.5e-07, "deprecation_date": "2026-10-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -46026,6 +46186,7 @@ "supports_vision": false }, "o3-pro": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -46073,6 +46234,7 @@ "supports_web_search": true }, "o3-pro-2025-06-10": { + "supports_sampling_params": false, "deprecation_date": "2026-12-11", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -46120,6 +46282,7 @@ "supports_web_search": true }, "o4-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_flex": 1.38e-07, "cache_read_input_token_cost_priority": 5e-07, @@ -46156,6 +46319,7 @@ "supports_web_search": true }, "o4-mini-deep-research": { + "supports_sampling_params": false, "cache_read_input_token_cost": 5e-07, "input_cost_per_token": 2e-06, "litellm_provider": "openai", @@ -46190,6 +46354,7 @@ "successor_model": "gpt-5.6-sol" }, "o4-mini-deep-research-2025-06-26": { + "supports_sampling_params": false, "deprecation_date": "2026-07-23", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", "successor_model": "gpt-5.6-sol", @@ -46224,6 +46389,7 @@ "supports_pdf_input": true }, "o4-mini-2025-04-16": { + "supports_sampling_params": false, "cache_read_input_token_cost": 2.75e-07, "cache_read_input_token_cost_flex": 1.38e-07, "cache_read_input_token_cost_priority": 5e-07, @@ -59821,6 +59987,7 @@ ] }, "gpt-5-search-api": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "input_cost_per_token": 1.25e-06, "litellm_provider": "openai", @@ -59849,6 +60016,7 @@ "supports_minimal_reasoning_effort": true }, "gpt-5-search-api-2025-10-14": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-07, "input_cost_per_token": 1.25e-06, "litellm_provider": "openai", @@ -70570,6 +70738,7 @@ "source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrock/current/us-east-1/index.json" }, "azure/us-gov/gpt-5.1": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.71875e-07, "default_reasoning_effort": "none", "input_cost_per_token": 1.71875e-06, @@ -70604,6 +70773,7 @@ "supports_vision": true }, "azure/us-gov/o3-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 7.57e-07, "input_cost_per_token": 1.513e-06, "litellm_provider": "azure", @@ -71257,6 +71427,7 @@ "source": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing" }, "gpt-5.5-cyber": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-06, "input_cost_per_token": 1.25e-05, "litellm_provider": "openai", @@ -71267,6 +71438,7 @@ "supports_reasoning": true }, "gpt-5.5-cyber-preview": { + "supports_sampling_params": false, "cache_read_input_token_cost": 1.25e-06, "deprecation_date": "2027-02-01", "deprecation_source": "https://developers.openai.com/api/docs/deprecations", @@ -71725,6 +71897,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.1", @@ -71747,6 +71920,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-03-17", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.1-codex", @@ -71758,6 +71932,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.4-mini", @@ -71780,6 +71955,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.4-nano", @@ -71799,6 +71975,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-04-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.2-pro", @@ -71816,6 +71993,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.1-codex-max": { + "supports_sampling_params": false, "deprecation_date": "2027-05-18", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.375e-07, @@ -71826,6 +72004,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.2": { + "supports_sampling_params": false, "deprecation_date": "2027-06-08", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.4", @@ -71848,6 +72027,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.2-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.3-codex", @@ -71859,6 +72039,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.31e-05, "input_cost_per_token_batches": 1.155e-05, "litellm_provider": "azure", @@ -71873,6 +72054,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.3-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-08-24", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.925e-07, @@ -71891,6 +72073,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.4-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 8.25e-08, @@ -71912,6 +72095,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.4-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.2e-08, @@ -71930,6 +72114,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/gpt-5.4-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-09-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "input_cost_per_token": 3.3e-05, @@ -72094,6 +72279,7 @@ "supports_xhigh_reasoning_effort": true }, "azure/eu/o1-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -72104,6 +72290,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o3": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -72123,6 +72310,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o3-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "successor_model": "azure/eu/gpt-5.6-sol", "cache_read_input_token_cost": 5.5e-07, @@ -72142,6 +72330,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o3-deep-research": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6-sol", @@ -72153,6 +72342,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o4-mini": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6-terra", @@ -72173,6 +72363,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/eu/o4-mini-2025-04-16": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/eu/gpt-5.6-terra", @@ -72349,6 +72540,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.1", @@ -72371,6 +72563,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-03-17", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.1-codex", @@ -72382,6 +72575,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.4-mini", @@ -72404,6 +72598,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-02-09", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.4-nano", @@ -72423,6 +72618,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-04-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.2-pro", @@ -72440,6 +72636,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.1-codex-max": { + "supports_sampling_params": false, "deprecation_date": "2027-05-18", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.375e-07, @@ -72450,6 +72647,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.2": { + "supports_sampling_params": false, "deprecation_date": "2027-06-08", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.4", @@ -72472,6 +72670,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.2-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-07-13", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.3-codex", @@ -72483,6 +72682,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.2-pro": { + "supports_sampling_params": false, "input_cost_per_token": 2.31e-05, "input_cost_per_token_batches": 1.155e-05, "litellm_provider": "azure", @@ -72497,6 +72697,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.3-codex": { + "supports_sampling_params": false, "deprecation_date": "2027-08-24", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 1.925e-07, @@ -72515,6 +72716,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.4-mini": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 8.25e-08, @@ -72536,6 +72738,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.4-nano": { + "supports_sampling_params": false, "deprecation_date": "2027-09-21", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "cache_read_input_token_cost": 2.2e-08, @@ -72554,6 +72757,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/gpt-5.4-pro": { + "supports_sampling_params": false, "deprecation_date": "2027-09-07", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "input_cost_per_token": 3.3e-05, @@ -72574,6 +72778,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/o1-mini": { + "supports_sampling_params": false, "cache_read_input_token_cost": 6.05e-07, "input_cost_per_token": 1.21e-06, "input_cost_per_token_batches": 6.05e-07, @@ -72584,6 +72789,7 @@ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'" }, "azure/us/o3-deep-research": { + "supports_sampling_params": false, "deprecation_date": "2026-11-19", "deprecation_source": "https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule", "successor_model": "azure/us/gpt-5.6-sol", diff --git a/tests/unit/llms/azure/response/test_azure_transformation.py b/tests/unit/llms/azure/response/test_azure_transformation.py index 05c0817e19b..38b0cdcd01a 100644 --- a/tests/unit/llms/azure/response/test_azure_transformation.py +++ b/tests/unit/llms/azure/response/test_azure_transformation.py @@ -1814,3 +1814,12 @@ def test_azure_reasoning_replay_drops_status_always_and_content_under_drop_param assert request["input"] == [replayable if drop_params else {**replayable, "content": saved_content}] assert history == original + + +@pytest.mark.parametrize("model", ["o3", "azure/o3", "azure/us/o3"]) +@pytest.mark.parametrize("param,value", [("temperature", 0.5), ("top_p", 0.9)]) +def test_azure_o_series_responses_sampling_cost_map(model: str, param: str, value: float) -> None: + config: Final = AzureOpenAIOSeriesResponsesAPIConfig() + assert config.map_openai_params({param: value}, model, drop_params=True) == {} + with pytest.raises(litellm.UnsupportedParamsError): + config.map_openai_params({param: value}, model, drop_params=False) diff --git a/tests/unit/llms/openai/responses/test_openai_responses_transformation.py b/tests/unit/llms/openai/responses/test_openai_responses_transformation.py index cd300d74b59..f0a444b3afe 100644 --- a/tests/unit/llms/openai/responses/test_openai_responses_transformation.py +++ b/tests/unit/llms/openai/responses/test_openai_responses_transformation.py @@ -13,6 +13,7 @@ from openai.types.responses.response_compaction_item import ResponseCompactionIt from pydantic import JsonValue, TypeAdapter import litellm + from litellm.litellm_core_utils.litellm_logging import Logging as LiteLLMLoggingObj from litellm.llms.azure.responses.transformation import AzureOpenAIResponsesAPIConfig from litellm.llms.azure_ai.responses.transformation import AzureAIResponsesAPIConfig @@ -5052,3 +5053,40 @@ def test_compact_request_keeps_stream_under_drop_params(config: OpenAIResponsesA ) assert {key: data[key] for key in data if key != "input"} == {"model": "gpt-5.4-mini", "stream": True} + + +@pytest.mark.parametrize("param,value", [("temperature", 0.5), ("top_p", 0.9)]) +@pytest.mark.parametrize("model", ["o3", "openai/o3", "o4-mini", "o1-pro"]) +def test_o_series_responses_rejects_or_drops_sampling(model: str, param: str, value: float) -> None: + config: Final = OpenAIResponsesAPIConfig() + assert config.map_openai_params({param: value}, model, drop_params=True) == {} + with pytest.raises(litellm.UnsupportedParamsError): + config.map_openai_params({param: value}, model, drop_params=False) + + +@pytest.mark.parametrize( + "model,reasoning,expected", + [ + ("gpt-5.1", {"effort": "none"}, {"temperature": 0.5, "top_p": 0.9, "reasoning": {"effort": "none"}}), + ("gpt-5.5", {"effort": "none"}, {"temperature": 0.5, "top_p": 0.9, "reasoning": {"effort": "none"}}), + ("gpt-5.5", {"effort": "medium"}, {"reasoning": {"effort": "medium"}}), + ("gpt-5-chat-latest", None, {"temperature": 0.5, "top_p": 0.9}), + ("openai/custom-sampling-deployment", None, {"temperature": 0.5, "top_p": 0.9}), + ("openai/gpt-5-custom-sampling-deployment", None, {}), + ], +) +def test_responses_sampling_effort_and_unknown_fallback( + model: str, reasoning: dict[str, str] | None, expected: dict[str, object] +) -> None: + params: Final = {"temperature": 0.5, "top_p": 0.9, **({"reasoning": reasoning} if reasoning else {})} + assert OpenAIResponsesAPIConfig().map_openai_params(params, model, drop_params=True) == expected + + +def test_responses_empty_sampling_params() -> None: + assert OpenAIResponsesAPIConfig().map_openai_params({}, "o3", drop_params=False) == {} + + +@pytest.mark.parametrize("drop_params", [True, False]) +def test_hosted_vllm_responses_keeps_sampling_when_effort_is_none(drop_params: bool) -> None: + params: Final = {"temperature": 0.5, "top_p": 0.9, "reasoning": {"effort": "none"}} + assert HostedVLLMResponsesAPIConfig().map_openai_params(params, "gpt-5.1", drop_params) == params diff --git a/tests/unit/llms/openai/test_o_series_transformation.py b/tests/unit/llms/openai/test_o_series_transformation.py index 9d39dfda60a..ff2899d13a7 100644 --- a/tests/unit/llms/openai/test_o_series_transformation.py +++ b/tests/unit/llms/openai/test_o_series_transformation.py @@ -10,6 +10,21 @@ from litellm.llms.openai.chat.o_series_transformation import OpenAIOSeriesConfig import os +@pytest.mark.parametrize("param,value", [("temperature", 0.5), ("top_p", 0.9)]) +def test_o_series_chat_sampling_cost_map_override(param: str, value: float) -> None: + model: Final = "openai/o3-sampling-enabled-test" + with patch.dict(litellm.model_cost, {model: {"litellm_provider": "openai", "mode": "chat", "supports_sampling_params": True}}): + assert OpenAIOSeriesConfig().map_openai_params({param: value}, {}, model, False) == {param: value} + + +@pytest.mark.parametrize("param,value", [("temperature", 0.5), ("top_p", 0.9)]) +def test_o_series_chat_sampling_unknown_fallback(param: str, value: float) -> None: + config: Final = OpenAIOSeriesConfig() + assert config.map_openai_params({param: value}, {}, "o3-unknown-test", True) == {} + with pytest.raises(litellm.UnsupportedParamsError): + config.map_openai_params({param: value}, {}, "o3-unknown-test", False) + + @pytest.mark.parametrize( "model_name,expected", [