BerriAI/litellm · error · ValueError

Invalid content type: {type(content)}

Error message

Invalid content type: {type(content)}

What it means

Raised while converting Responses API input items to chat-completion content when a content block has an unrecognized type; the request content object shape is not one of the supported text/input_text/output_text/image variants.

Source

Thrown at litellm/responses/litellm_completion_transformation/transformation.py:1351

                                OpenAIChatCompletionTextObject(type="text", text=str(encrypted_content))
                            )
                    else:
                        # Skip text blocks with None text to avoid downstream errors
                        text_value = item.get("text")
                        if text_value is None:
                            continue
                        content_block: dict[str, object] = {
                            "type": LiteLLMCompletionResponsesConfig._get_chat_completion_request_content_type(
                                item.get("type") or "text"
                            ),
                            "text": text_value,
                        }
                        if "cache_control" in item:
                            content_block["cache_control"] = item["cache_control"]
                        content_list.append(content_block)
            return content_list
        else:
            raise ValueError(f"Invalid content type: {type(content)}")

    @staticmethod
    def _get_chat_completion_request_content_type(
        content_type: str,
    ) -> ValidChatCompletionMessageContentTypesLiteral:
        """
        Transform Responses API content type to valid Chat Completion content type.

        Returns one of ValidChatCompletionMessageContentTypes:
        - User: "text", "image_url", "input_audio", "audio_url", "document",
                "guarded_text", "video_url", "file"
        - Assistant: "text", "thinking", "redacted_thinking"
        """
        # Responses API content has `input_` prefix, if it exists, remove it
        if content_type.startswith("input_"):
            stripped: Final = content_type[len("input_") :]
            # Validate stripped type is valid, otherwise default to "text"
            if stripped in ValidChatCompletionMessageContentTypes:

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Pass content as a string or a supported content-parts list.

When it happens

Trigger: Thrown at litellm/responses/litellm_completion_transformation/transformation.py:1351 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/407008140f73705c. Report an issue: GitHub.