BerriAI/litellm · error · ValueError

litellm_params is required

Error message

litellm_params is required

What it means

validate_input_kwargs raises when 'litellm_params' is absent from kwargs or is not a dict. litellm_params carries router/proxy-level metadata (metadata, base_url, api_base, ...) that the bridge threads into the Responses API call. Only direct, hand-built invocations of the handler can omit it.

Source

Thrown at litellm/completion_extras/litellm_responses_transformation/handler.py:126

        model: Final = typed_kwargs.get("model")
        if model is None or not isinstance(model, str):
            raise ValueError("model is required")

        custom_llm_provider: Final = typed_kwargs.get("custom_llm_provider")
        if custom_llm_provider is None or not isinstance(custom_llm_provider, str):
            raise ValueError("custom_llm_provider is required")

        messages: Final = typed_kwargs.get("messages")
        if messages is None or not isinstance(messages, list):
            raise ValueError("messages is required")

        optional_params: Final = typed_kwargs.get("optional_params")
        if optional_params is None or not isinstance(optional_params, dict):
            raise ValueError("optional_params is required")

        litellm_params: Final = typed_kwargs.get("litellm_params")
        if litellm_params is None or not isinstance(litellm_params, dict):
            raise ValueError("litellm_params is required")

        headers: Final = typed_kwargs.get("headers")
        if headers is None or not isinstance(headers, dict):
            raise ValueError("headers is required")

        model_response: Final = typed_kwargs.get("model_response")
        if model_response is None or not isinstance(model_response, ModelResponse):
            raise ValueError("model_response is required")

        logging_obj: Final = typed_kwargs.get("logging_obj")
        if logging_obj is None or not isinstance(logging_obj, LiteLLMLoggingObj):
            raise ValueError("logging_obj is required")

        return ResponsesToCompletionBridgeHandlerInputKwargs(
            model=model,
            messages=messages,
            optional_params=optional_params,
            litellm_params=litellm_params,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Use litellm.completion()/acompletion() which always injects litellm_params
2. If direct, include litellm_params as a dict (an empty dict is accepted)
3. Verify no middleware strips the key before it reaches the handler

Example fix

# before
kwargs = {"model": m, "messages": msgs, "optional_params": {}}

# after
kwargs = {"model": m, "messages": msgs, "optional_params": {}, "litellm_params": {}}

def has_litellm_params(kwargs: dict) -> bool:
    return isinstance(kwargs.get("litellm_params"), dict)

def is_dict_param(v) -> bool:
    return isinstance(v, dict)

When it happens

Trigger: Calling the bridge handler with a kwargs dict missing the litellm_params key; passing it as a non-dict (e.g. a list of tuples).

Common situations: Custom pipelines or unit tests constructing kwargs manually; refactors that pass only the outer request params and forget the litellm-internal dict.

Related errors

• model is required
• custom_llm_provider is required
• messages is required
• optional_params is required
• headers is required

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/fb249611fb29058b. Report an issue: GitHub.