Merge pull request #9185 from BerriAI/litellm_dev_03_12_2025_p1

Litellm dev 03 12 2025 p1

litellm/litellm_core_utils/get_litellm_params.py

@@ -58,6 +58,7 @@ def get_litellm_params(
     async_call: Optional[bool] = None,
     ssl_verify: Optional[bool] = None,
     merge_reasoning_content_in_choices: Optional[bool] = None,
+    api_version: Optional[str] = None,
     max_retries: Optional[int] = None,
     **kwargs,
 ) -> dict:
@@ -100,6 +101,7 @@ def get_litellm_params(
         "async_call": async_call,
         "ssl_verify": ssl_verify,
         "merge_reasoning_content_in_choices": merge_reasoning_content_in_choices,
+        "api_version": api_version,
         "azure_ad_token": kwargs.get("azure_ad_token"),
         "tenant_id": kwargs.get("tenant_id"),
         "client_id": kwargs.get("client_id"),

litellm/llms/aiohttp_openai/chat/transformation.py

@@ -29,6 +29,7 @@ class AiohttpOpenAIChatConfig(OpenAILikeChatConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/azure_ai/chat/transformation.py

@@ -67,6 +67,7 @@ class AzureAIStudioConfig(OpenAIConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """
@@ -92,12 +93,14 @@ class AzureAIStudioConfig(OpenAIConfig):
         original_url = httpx.URL(api_base)
 
         # Extract api_version or use default
-        api_version = cast(Optional[str], optional_params.get("api_version"))
+        api_version = cast(Optional[str], litellm_params.get("api_version"))
 
-        # Check if 'api-version' is already present
-        if "api-version" not in original_url.params and api_version:
-            # Add api_version to optional_params
-            original_url.params["api-version"] = api_version
+        # Create a new dictionary with existing params
+        query_params = dict(original_url.params)
+
+        # Add api_version if needed
+        if "api-version" not in query_params and api_version:
+            query_params["api-version"] = api_version
 
         # Add the path to the base URL
         if "services.ai.azure.com" in api_base:
@@ -109,8 +112,7 @@ class AzureAIStudioConfig(OpenAIConfig):
                 api_base=api_base, ending_path="/chat/completions"
             )
 
-        # Convert optional_params to query parameters
-        query_params = original_url.params
+        # Use the new query_params dictionary
         final_url = httpx.URL(new_url).copy_with(params=query_params)
 
         return str(final_url)

litellm/llms/base_llm/audio_transcription/transformation.py

@@ -30,6 +30,7 @@ class BaseAudioTranscriptionConfig(BaseConfig, ABC):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/base_llm/chat/transformation.py

@@ -270,6 +270,7 @@ class BaseConfig(ABC):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/base_llm/completion/transformation.py

@@ -31,6 +31,7 @@ class BaseTextCompletionConfig(BaseConfig, ABC):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/base_llm/embedding/transformation.py

@@ -45,6 +45,7 @@ class BaseEmbeddingConfig(BaseConfig, ABC):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/base_llm/image_variations/transformation.py

@@ -36,6 +36,7 @@ class BaseImageVariationConfig(BaseConfig, ABC):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/bedrock/chat/invoke_transformations/base_invoke_transformation.py

@@ -76,6 +76,7 @@ class AmazonInvokeConfig(BaseConfig, BaseAWSLLM):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/cloudflare/chat/transformation.py

@@ -79,6 +79,7 @@ class CloudflareChatConfig(BaseConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         if api_base is None:

litellm/llms/custom_httpx/aiohttp_handler.py

@@ -234,6 +234,7 @@ class BaseLLMAIOHTTPHandler:
             api_base=api_base,
             model=model,
             optional_params=optional_params,
+            litellm_params=litellm_params,
             stream=stream,
         )
 
@@ -483,6 +484,7 @@ class BaseLLMAIOHTTPHandler:
             api_base=api_base,
             model=model,
             optional_params=optional_params,
+            litellm_params=litellm_params,
             stream=False,
         )
 

litellm/llms/custom_httpx/llm_http_handler.py

@@ -242,6 +242,7 @@ class BaseLLMHTTPHandler:
             model=model,
             optional_params=optional_params,
             stream=stream,
+            litellm_params=litellm_params,
         )
 
         data = provider_config.transform_request(
@@ -612,6 +613,7 @@ class BaseLLMHTTPHandler:
             api_base=api_base,
             model=model,
             optional_params=optional_params,
+            litellm_params=litellm_params,
         )
 
         data = provider_config.transform_embedding_request(
@@ -907,6 +909,7 @@ class BaseLLMHTTPHandler:
         client: Optional[Union[HTTPHandler, AsyncHTTPHandler]] = None,
         atranscription: bool = False,
         headers: dict = {},
+        litellm_params: dict = {},
     ) -> TranscriptionResponse:
         provider_config = ProviderConfigManager.get_provider_audio_transcription_config(
             model=model, provider=litellm.LlmProviders(custom_llm_provider)
@@ -930,6 +933,7 @@ class BaseLLMHTTPHandler:
             api_base=api_base,
             model=model,
             optional_params=optional_params,
+            litellm_params=litellm_params,
         )
 
         # Handle the audio file based on type

litellm/llms/deepgram/audio_transcription/transformation.py

@@ -103,6 +103,7 @@ class DeepgramAudioTranscriptionConfig(BaseAudioTranscriptionConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         if api_base is None:

litellm/llms/deepseek/chat/transformation.py

@@ -40,6 +40,7 @@ class DeepSeekChatConfig(OpenAIGPTConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/ollama/completion/transformation.py

@@ -356,6 +356,7 @@ class OllamaConfig(BaseConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/openai/chat/gpt_transformation.py

@@ -291,6 +291,7 @@ class OpenAIGPTConfig(BaseLLMModelInfo, BaseConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         """

litellm/llms/openai_like/chat/handler.py

@@ -230,7 +230,7 @@ class OpenAILikeChatHandler(OpenAILikeBase):
         logging_obj,
         optional_params: dict,
         acompletion=None,
-        litellm_params=None,
+        litellm_params: dict = {},
         logger_fn=None,
         headers: Optional[dict] = None,
         timeout: Optional[Union[float, httpx.Timeout]] = None,
@@ -337,7 +337,7 @@ class OpenAILikeChatHandler(OpenAILikeBase):
                     timeout=timeout,
                     base_model=base_model,
                     client=client,
-                    json_mode=json_mode
+                    json_mode=json_mode,
                 )
         else:
             ## COMPLETION CALL

litellm/llms/replicate/chat/handler.py

@@ -169,7 +169,10 @@ def completion(
     )  # for pricing this must remain right before calling api
 
     prediction_url = replicate_config.get_complete_url(
-        api_base=api_base, model=model, optional_params=optional_params
+        api_base=api_base,
+        model=model,
+        optional_params=optional_params,
+        litellm_params=litellm_params,
     )
 
     ## COMPLETION CALL
@@ -243,7 +246,10 @@ async def async_completion(
 ) -> Union[ModelResponse, CustomStreamWrapper]:
 
     prediction_url = replicate_config.get_complete_url(
-        api_base=api_base, model=model, optional_params=optional_params
+        api_base=api_base,
+        model=model,
+        optional_params=optional_params,
+        litellm_params=litellm_params,
     )
     async_handler = get_async_httpx_client(
         llm_provider=litellm.LlmProviders.REPLICATE,

litellm/llms/replicate/chat/transformation.py

@@ -141,6 +141,7 @@ class ReplicateConfig(BaseConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         version_id = self.model_to_version_id(model)

litellm/llms/topaz/image_variations/transformation.py

@@ -55,6 +55,7 @@ class TopazImageVariationConfig(BaseImageVariationConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         api_base = api_base or "https://api.topazlabs.com"

litellm/llms/triton/completion/transformation.py

@@ -72,6 +72,7 @@ class TritonConfig(BaseConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         if api_base is None:

litellm/llms/voyage/embedding/transformation.py

@@ -43,6 +43,7 @@ class VoyageEmbeddingConfig(BaseEmbeddingConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         if api_base:

litellm/llms/watsonx/chat/handler.py

@@ -31,7 +31,7 @@ class WatsonXChatHandler(OpenAILikeChatHandler):
         logging_obj,
         optional_params: dict,
         acompletion=None,
-        litellm_params=None,
+        litellm_params: dict = {},
         headers: Optional[dict] = None,
         logger_fn=None,
         timeout: Optional[Union[float, httpx.Timeout]] = None,
@@ -63,6 +63,7 @@ class WatsonXChatHandler(OpenAILikeChatHandler):
             api_base=api_base,
             model=model,
             optional_params=optional_params,
+            litellm_params=litellm_params,
             stream=optional_params.get("stream", False),
         )
 

litellm/llms/watsonx/chat/transformation.py

@@ -83,6 +83,7 @@ class IBMWatsonXChatConfig(IBMWatsonXMixin, OpenAIGPTConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         url = self._get_base_url(api_base=api_base)

litellm/llms/watsonx/completion/transformation.py

@@ -318,6 +318,7 @@ class IBMWatsonXAIConfig(IBMWatsonXMixin, BaseConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         url = self._get_base_url(api_base=api_base)

litellm/llms/watsonx/embed/transformation.py

@@ -54,6 +54,7 @@ class IBMWatsonXEmbeddingConfig(IBMWatsonXMixin, BaseEmbeddingConfig):
         api_base: Optional[str],
         model: str,
         optional_params: dict,
+        litellm_params: dict,
         stream: Optional[bool] = None,
     ) -> str:
         url = self._get_base_url(api_base=api_base)

litellm/main.py

@@ -1162,6 +1162,7 @@ def completion(  # type: ignore # noqa: PLR0915
             merge_reasoning_content_in_choices=kwargs.get(
                 "merge_reasoning_content_in_choices", None
             ),
+            api_version=api_version,
             azure_ad_token=kwargs.get("azure_ad_token"),
             tenant_id=kwargs.get("tenant_id"),
             client_id=kwargs.get("client_id"),

litellm/model_prices_and_context_window_backup.json

@@ -1021,6 +1021,120 @@
         "input_cost_per_character": 0.000030,
         "litellm_provider": "openai"
     },
+    "azure/gpt-4o-mini-realtime-preview-2024-12-17": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.0000006,
+        "input_cost_per_audio_token": 0.00001,
+        "cache_read_input_token_cost": 0.0000003,
+        "cache_creation_input_audio_token_cost": 0.0000003,
+        "output_cost_per_token": 0.0000024,
+        "output_cost_per_audio_token": 0.00002,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-mini-realtime-preview-2024-12-17": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.00000066,
+        "input_cost_per_audio_token": 0.000011,
+        "cache_read_input_token_cost": 0.00000033,
+        "cache_creation_input_audio_token_cost": 0.00000033,
+        "output_cost_per_token": 0.00000264,
+        "output_cost_per_audio_token": 0.000022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/us/gpt-4o-mini-realtime-preview-2024-12-17": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.00000066,
+        "input_cost_per_audio_token": 0.000011,
+        "cache_read_input_token_cost": 0.00000033,
+        "cache_creation_input_audio_token_cost": 0.00000033,
+        "output_cost_per_token": 0.00000264,
+        "output_cost_per_audio_token": 0.000022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/gpt-4o-realtime-preview-2024-10-01": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000005,
+        "input_cost_per_audio_token": 0.0001,
+        "cache_read_input_token_cost": 0.0000025,
+        "cache_creation_input_audio_token_cost": 0.00002,
+        "output_cost_per_token": 0.00002,
+        "output_cost_per_audio_token": 0.0002,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/us/gpt-4o-realtime-preview-2024-10-01": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.0000055,
+        "input_cost_per_audio_token": 0.00011,
+        "cache_read_input_token_cost": 0.00000275,
+        "cache_creation_input_audio_token_cost": 0.000022,
+        "output_cost_per_token": 0.000022,
+        "output_cost_per_audio_token": 0.00022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-realtime-preview-2024-10-01": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.0000055,
+        "input_cost_per_audio_token": 0.00011,
+        "cache_read_input_token_cost": 0.00000275,
+        "cache_creation_input_audio_token_cost": 0.000022,
+        "output_cost_per_token": 0.000022,
+        "output_cost_per_audio_token": 0.00022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
     "azure/o3-mini-2025-01-31": {
         "max_tokens": 100000,
         "max_input_tokens": 200000,
@@ -1034,6 +1148,36 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/o3-mini-2025-01-31": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_vision": false,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/o3-mini-2025-01-31": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_vision": false,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/tts-1": {
         "mode": "audio_speech", 
         "input_cost_per_character": 0.000015,
@@ -1092,6 +1236,38 @@
         "supports_vision": false,
         "supports_prompt_caching": true
     },
+    "azure/us/o1-mini-2024-09-12": {
+        "max_tokens": 65536,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 65536,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
+    "azure/eu/o1-mini-2024-09-12": {
+        "max_tokens": 65536,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 65536,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
     "azure/o1": {
         "max_tokens": 100000,
         "max_input_tokens": 200000,
@@ -1122,6 +1298,36 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/o1-2024-12-17": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/o1-2024-12-17": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/o1-preview": {
         "max_tokens": 32768,
         "max_input_tokens": 128000,
@@ -1150,6 +1356,34 @@
         "supports_vision": false,
         "supports_prompt_caching": true
     },
+    "azure/us/o1-preview-2024-09-12": {
+        "max_tokens": 32768,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 32768,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
+    "azure/eu/o1-preview-2024-09-12": {
+        "max_tokens": 32768,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 32768,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
     "azure/gpt-4o": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -1195,6 +1429,36 @@
         "supports_vision": true,
         "supports_tool_choice": true
     },
+    "azure/us/gpt-4o-2024-11-20": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "cache_creation_input_token_cost": 0.00000138,
+        "output_cost_per_token": 0.000011,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-2024-11-20": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "cache_creation_input_token_cost": 0.00000138,
+        "output_cost_per_token": 0.000011,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_tool_choice": true
+    },
     "azure/gpt-4o-2024-05-13": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -1225,6 +1489,38 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/gpt-4o-2024-08-06": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "output_cost_per_token": 0.000011,
+        "cache_read_input_token_cost": 0.000001375,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-2024-08-06": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "output_cost_per_token": 0.000011,
+        "cache_read_input_token_cost": 0.000001375,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/global-standard/gpt-4o-2024-11-20": {
         "max_tokens": 16384,
         "max_input_tokens": 128000,
@@ -1285,6 +1581,38 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/gpt-4o-mini-2024-07-18": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000000165,
+        "output_cost_per_token": 0.00000066,
+        "cache_read_input_token_cost": 0.000000083,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-mini-2024-07-18": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000000165,
+        "output_cost_per_token": 0.00000066,
+        "cache_read_input_token_cost": 0.000000083,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/gpt-4-turbo-2024-04-09": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,

litellm/proxy/_new_secret_config.yaml

@@ -14,5 +14,4 @@ credential_list:
       description: "Azure API Key and Base URL"
       type: "azure"
       required: true
-      default: "azure"
-
+      default: "azure"

model_prices_and_context_window.json

@@ -1021,6 +1021,120 @@
         "input_cost_per_character": 0.000030,
         "litellm_provider": "openai"
     },
+    "azure/gpt-4o-mini-realtime-preview-2024-12-17": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.0000006,
+        "input_cost_per_audio_token": 0.00001,
+        "cache_read_input_token_cost": 0.0000003,
+        "cache_creation_input_audio_token_cost": 0.0000003,
+        "output_cost_per_token": 0.0000024,
+        "output_cost_per_audio_token": 0.00002,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-mini-realtime-preview-2024-12-17": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.00000066,
+        "input_cost_per_audio_token": 0.000011,
+        "cache_read_input_token_cost": 0.00000033,
+        "cache_creation_input_audio_token_cost": 0.00000033,
+        "output_cost_per_token": 0.00000264,
+        "output_cost_per_audio_token": 0.000022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/us/gpt-4o-mini-realtime-preview-2024-12-17": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.00000066,
+        "input_cost_per_audio_token": 0.000011,
+        "cache_read_input_token_cost": 0.00000033,
+        "cache_creation_input_audio_token_cost": 0.00000033,
+        "output_cost_per_token": 0.00000264,
+        "output_cost_per_audio_token": 0.000022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/gpt-4o-realtime-preview-2024-10-01": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.000005,
+        "input_cost_per_audio_token": 0.0001,
+        "cache_read_input_token_cost": 0.0000025,
+        "cache_creation_input_audio_token_cost": 0.00002,
+        "output_cost_per_token": 0.00002,
+        "output_cost_per_audio_token": 0.0002,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/us/gpt-4o-realtime-preview-2024-10-01": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.0000055,
+        "input_cost_per_audio_token": 0.00011,
+        "cache_read_input_token_cost": 0.00000275,
+        "cache_creation_input_audio_token_cost": 0.000022,
+        "output_cost_per_token": 0.000022,
+        "output_cost_per_audio_token": 0.00022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-realtime-preview-2024-10-01": {
+        "max_tokens": 4096,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 4096,
+        "input_cost_per_token": 0.0000055,
+        "input_cost_per_audio_token": 0.00011,
+        "cache_read_input_token_cost": 0.00000275,
+        "cache_creation_input_audio_token_cost": 0.000022,
+        "output_cost_per_token": 0.000022,
+        "output_cost_per_audio_token": 0.00022,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_audio_input": true,
+        "supports_audio_output": true,
+        "supports_system_messages": true,
+        "supports_tool_choice": true
+    },
     "azure/o3-mini-2025-01-31": {
         "max_tokens": 100000,
         "max_input_tokens": 200000,
@@ -1034,6 +1148,36 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/o3-mini-2025-01-31": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_vision": false,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/o3-mini-2025-01-31": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_vision": false,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/tts-1": {
         "mode": "audio_speech", 
         "input_cost_per_character": 0.000015,
@@ -1092,6 +1236,38 @@
         "supports_vision": false,
         "supports_prompt_caching": true
     },
+    "azure/us/o1-mini-2024-09-12": {
+        "max_tokens": 65536,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 65536,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
+    "azure/eu/o1-mini-2024-09-12": {
+        "max_tokens": 65536,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 65536,
+        "input_cost_per_token": 0.00000121,
+        "input_cost_per_token_batches": 0.000000605,
+        "output_cost_per_token": 0.00000484,
+        "output_cost_per_token_batches": 0.00000242,
+        "cache_read_input_token_cost": 0.000000605,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
     "azure/o1": {
         "max_tokens": 100000,
         "max_input_tokens": 200000,
@@ -1122,6 +1298,36 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/o1-2024-12-17": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/o1-2024-12-17": {
+        "max_tokens": 100000,
+        "max_input_tokens": 200000,
+        "max_output_tokens": 100000,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/o1-preview": {
         "max_tokens": 32768,
         "max_input_tokens": 128000,
@@ -1150,6 +1356,34 @@
         "supports_vision": false,
         "supports_prompt_caching": true
     },
+    "azure/us/o1-preview-2024-09-12": {
+        "max_tokens": 32768,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 32768,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
+    "azure/eu/o1-preview-2024-09-12": {
+        "max_tokens": 32768,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 32768,
+        "input_cost_per_token": 0.0000165,
+        "output_cost_per_token": 0.000066,
+        "cache_read_input_token_cost": 0.00000825,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_vision": false,
+        "supports_prompt_caching": true
+    },
     "azure/gpt-4o": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -1195,6 +1429,36 @@
         "supports_vision": true,
         "supports_tool_choice": true
     },
+    "azure/us/gpt-4o-2024-11-20": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "cache_creation_input_token_cost": 0.00000138,
+        "output_cost_per_token": 0.000011,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-2024-11-20": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "cache_creation_input_token_cost": 0.00000138,
+        "output_cost_per_token": 0.000011,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_tool_choice": true
+    },
     "azure/gpt-4o-2024-05-13": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,
@@ -1225,6 +1489,38 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/gpt-4o-2024-08-06": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "output_cost_per_token": 0.000011,
+        "cache_read_input_token_cost": 0.000001375,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-2024-08-06": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.00000275,
+        "output_cost_per_token": 0.000011,
+        "cache_read_input_token_cost": 0.000001375,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/global-standard/gpt-4o-2024-11-20": {
         "max_tokens": 16384,
         "max_input_tokens": 128000,
@@ -1285,6 +1581,38 @@
         "supports_prompt_caching": true,
         "supports_tool_choice": true
     },
+    "azure/us/gpt-4o-mini-2024-07-18": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000000165,
+        "output_cost_per_token": 0.00000066,
+        "cache_read_input_token_cost": 0.000000083,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
+    "azure/eu/gpt-4o-mini-2024-07-18": {
+        "max_tokens": 16384,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "input_cost_per_token": 0.000000165,
+        "output_cost_per_token": 0.00000066,
+        "cache_read_input_token_cost": 0.000000083,
+        "litellm_provider": "azure",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "supports_parallel_function_calling": true,
+        "supports_response_schema": true,
+        "supports_vision": true,
+        "supports_prompt_caching": true,
+        "supports_tool_choice": true
+    },
     "azure/gpt-4-turbo-2024-04-09": {
         "max_tokens": 4096,
         "max_input_tokens": 128000,

tests/litellm_utils_tests/test_supports_tool_choice.py

@@ -117,7 +117,9 @@ async def test_supports_tool_choice():
     """
     # Load model prices
     litellm._turn_on_debug()
-    with open("./model_prices_and_context_window.json", "r") as f:
+    local_path = "../../model_prices_and_context_window.json"
+    prod_path = "./model_prices_and_context_window.json"
+    with open(prod_path, "r") as f:
         model_prices = json.load(f)
     litellm.model_cost = model_prices
     config_manager = ProviderConfigManager()
@@ -133,6 +135,8 @@ async def test_supports_tool_choice():
             or any(provider in model_name for provider in OLD_PROVIDERS)
             or model_info["litellm_provider"] in OLD_PROVIDERS
             or model_name in block_list
+            or "azure/eu" in model_name
+            or "azure/us" in model_name
         ):
             continue
 

tests/llm_translation/test_azure_ai.py

@@ -159,6 +159,32 @@ def test_azure_ai_services_handler(api_base, expected_url):
         assert mock_client.call_args.kwargs["url"] == expected_url
 
 
+def test_azure_ai_services_with_api_version():
+    from litellm.llms.custom_httpx.http_handler import HTTPHandler, AsyncHTTPHandler
+
+    client = HTTPHandler()
+
+    with patch.object(client, "post") as mock_client:
+        try:
+            response = litellm.completion(
+                model="azure_ai/Meta-Llama-3.1-70B-Instruct",
+                messages=[{"role": "user", "content": "Hello, how are you?"}],
+                api_key="my-fake-api-key",
+                api_version="2024-05-01-preview",
+                api_base="https://litellm8397336933.services.ai.azure.com/models",
+                client=client,
+            )
+        except Exception as e:
+            print(f"Error: {e}")
+
+        mock_client.assert_called_once()
+        assert mock_client.call_args.kwargs["headers"]["api-key"] == "my-fake-api-key"
+        assert (
+            mock_client.call_args.kwargs["url"]
+            == "https://litellm8397336933.services.ai.azure.com/models/chat/completions?api-version=2024-05-01-preview"
+        )
+
+
 def test_completion_azure_ai_command_r():
     try:
         import os

tests/llm_translation/test_unit_test_bedrock_invoke.py

@@ -28,6 +28,7 @@ def test_get_complete_url_basic(bedrock_transformer):
         model="anthropic.claude-v2",
         optional_params={},
         stream=False,
+        litellm_params={},
     )
 
     assert (
@@ -43,6 +44,7 @@ def test_get_complete_url_streaming(bedrock_transformer):
         model="anthropic.claude-v2",
         optional_params={},
         stream=True,
+        litellm_params={},
     )
 
     assert (