fix: don't fail request if unmapped item in responses list

not every responses item has a 1:1 mapping with chat completions

litellm/completion_extras/litellm_responses_transformation/transformation.py

@@ -277,13 +277,8 @@ class LiteLLMResponsesTransformationHandler(CompletionTransformationBridge):
                     Choices(message=msg, finish_reason="tool_calls", index=index)
                 )
                 index += 1
-            elif isinstance(item, GenericResponseOutputItem):
-                raise ValueError("GenericResponseOutputItem not supported")
-            elif isinstance(item, OutputFunctionToolCall):
-                # function/tool calls pass through as-is
-                raise ValueError("Function calling not supported yet.")
             else:
-                raise ValueError(f"Unknown item type: {item}")
+                pass  # don't fail request if item in list is not supported
 
         if len(choices) == 0:
             if (
@@ -293,6 +288,10 @@ class LiteLLMResponsesTransformationHandler(CompletionTransformationBridge):
                 raise ValueError(
                     f"{model} unable to complete request: {raw_response.incomplete_details.reason}"
                 )
+            else:
+                raise ValueError(
+                    f"Unknown items in responses API response: {raw_response.output}"
+                )
 
         setattr(model_response, "choices", choices)
 

tests/llm_translation/test_openai.py

@@ -552,7 +552,7 @@ def test_openai_deepresearch_model_bridge():
     litellm._turn_on_debug()
     response = litellm.completion(
         model="o3-deep-research-2025-06-26",
-        messages=[{"role": "user", "content": "What is the capital of France?"}],
+        messages=[{"role": "user", "content": "Hey, how's it going?"}],
         tools=[
             {"type": "web_search_preview"},
             {"type": "code_interpreter", "container": {"type": "auto"}},