Add exception mapping for responses API

litellm/litellm_core_utils/exception_mapping_utils.py

@@ -128,7 +128,6 @@ def exception_type(  # type: ignore  # noqa: PLR0915
     extra_kwargs: Optional[dict] = {},
 ):
     """Maps an LLM Provider Exception to OpenAI Exception Format"""
-
     if any(
         isinstance(original_exception, exc_type)
         for exc_type in litellm.LITELLM_EXCEPTION_TYPES

litellm/responses/main.py

@@ -58,15 +58,24 @@ async def aresponses(
     extra_query: Optional[Dict[str, Any]] = None,
     extra_body: Optional[Dict[str, Any]] = None,
     timeout: Optional[Union[float, httpx.Timeout]] = None,
+    # LiteLLM specific params,
+    custom_llm_provider: Optional[str] = None,
     **kwargs,
 ) -> Union[ResponsesAPIResponse, BaseResponsesAPIStreamingIterator]:
     """
     Async: Handles responses API requests by reusing the synchronous function
     """
+    local_vars = locals()
     try:
         loop = asyncio.get_event_loop()
         kwargs["aresponses"] = True
 
+        # get custom llm provider so we can use this for mapping exceptions
+        if custom_llm_provider is None:
+            _, custom_llm_provider, _, _ = litellm.get_llm_provider(
+                model=model, api_base=local_vars.get("base_url", None)
+            )
+
         func = partial(
             responses,
             input=input,
@@ -91,6 +100,7 @@ async def aresponses(
             extra_query=extra_query,
             extra_body=extra_body,
             timeout=timeout,
+            custom_llm_provider=custom_llm_provider,
             **kwargs,
         )
 
@@ -104,7 +114,13 @@ async def aresponses(
             response = init_response
         return response
     except Exception as e:
-        raise e
+        raise litellm.exception_type(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            original_exception=e,
+            completion_kwargs=local_vars,
+            extra_kwargs=kwargs,
+        )
 
 
 @client
@@ -133,12 +149,16 @@ def responses(
     extra_query: Optional[Dict[str, Any]] = None,
     extra_body: Optional[Dict[str, Any]] = None,
     timeout: Optional[Union[float, httpx.Timeout]] = None,
+    # LiteLLM specific params,
+    custom_llm_provider: Optional[str] = None,
     **kwargs,
 ):
     """
     Synchronous version of the Responses API.
     Uses the synchronous HTTP handler to make requests.
     """
+    local_vars = locals()
+    try:
         litellm_logging_obj: LiteLLMLoggingObj = kwargs.get("litellm_logging_obj")  # type: ignore
         litellm_call_id: Optional[str] = kwargs.get("litellm_call_id", None)
         _is_async = kwargs.pop("aresponses", False) is True
@@ -148,7 +168,7 @@ def responses(
         model, custom_llm_provider, dynamic_api_key, dynamic_api_base = (
             litellm.get_llm_provider(
                 model=model,
-            custom_llm_provider=kwargs.get("custom_llm_provider", None),
+                custom_llm_provider=custom_llm_provider,
                 api_base=litellm_params.api_base,
                 api_key=litellm_params.api_key,
             )
@@ -169,12 +189,12 @@ def responses(
                 message=f"Responses API not available for custom_llm_provider={custom_llm_provider}, model: {model}",
             )
 
-    # Get all parameters using locals() and combine with kwargs
-    local_vars = locals()
         local_vars.update(kwargs)
         # Get ResponsesAPIOptionalRequestParams with only valid parameters
         response_api_optional_params: ResponsesAPIOptionalRequestParams = (
-        ResponsesAPIRequestUtils.get_requested_response_api_optional_param(local_vars)
+            ResponsesAPIRequestUtils.get_requested_response_api_optional_param(
+                local_vars
+            )
         )
 
         # Get optional parameters for the responses API
@@ -215,3 +235,11 @@ def responses(
         )
 
         return response
+    except Exception as e:
+        raise litellm.exception_type(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            original_exception=e,
+            completion_kwargs=local_vars,
+            extra_kwargs=kwargs,
+        )