BerriAI/litellm · error · ValueError

model param not passed in.

Error message

model param not passed in.

What it means

Plain ValueError raised as the very first validation step of completion(): the model argument is None. Everything downstream (provider routing, auth, params) derives from the model string, so litellm refuses immediately instead of guessing.

Source

Thrown at litellm/main.py:4996

        api_key (str, optional): API key (default is None).
        model_list (list, optional): List of api base, version, keys
        extra_headers (dict, optional): Additional headers to include in the request.

        LITELLM Specific Params
        mock_response (str, optional): If provided, return a mock completion response for testing or debugging purposes (default is None).
        custom_llm_provider (str, optional): Used for Non-OpenAI LLMs, Example usage for bedrock, set model="amazon.titan-tg1-large" and custom_llm_provider="bedrock"
        max_retries (int, optional): The number of retries to attempt (default is 0).
    Returns:
        ModelResponse: A response object containing the generated completion and associated metadata.

    Note:
        - This function is used to perform completions() using the specified language model.
        - It supports various optional parameters for customizing the completion behavior.
        - If 'mock_response' is provided, a mock completion response is returned for testing or debugging.
    """
    ### VALIDATE Request ###
    if model is None:
        raise ValueError("model param not passed in.")
    # validate messages
    messages = validate_and_fix_openai_messages(messages=messages)
    tools = validate_and_fix_openai_tools(tools=tools)
    # validate tool_choice
    tool_choice = validate_chat_completion_tool_choice(tool_choice=tool_choice)
    # validate optional params
    stop = validate_openai_optional_params(stop=stop)
    # normalize camelCase thinking keys (e.g. budgetTokens -> budget_tokens)
    thinking = validate_and_fix_thinking_param(thinking=thinking)

    ######### unpacking kwargs #####################
    args: Final = _locals_snapshot(locals())

    # Set by the responses->completion fallback so completion() does not bridge
    # back to the Responses API: that round-trip mutually recurses forever for a
    # model whose model_cost mode is "responses" but whose provider has no
    # Responses API config (get_provider_responses_api_config -> None).
    skip_responses_api_bridge: Final = kwargs.pop("_skip_responses_api_bridge", False)

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Pass a concrete model string, e.g. 'gpt-4o' or 'azure/my-deploy'
2. Find why the variable is None: print it / assert before the call -- usually a missing env var or config key
3. Add a default at the source: model = os.environ.get('MODEL') or 'gpt-4o-mini'
4. Guard the call site with a quick isinstance(model, str) and bool(model.strip()) check

Example fix

# before
model = os.environ.get('MODEL_NAME')  # None when unset
resp = litellm.completion(model=model, messages=m)  # ValueError: model param not passed in.

# after
model = os.environ.get('MODEL_NAME') or 'gpt-4o-mini'
resp = litellm.completion(model=model, messages=m)

model = os.environ.get('MODEL') or DEFAULT_MODEL
assert isinstance(model, str) and model.strip(), 'model must resolve to a non-empty string'

from typing import TypeGuard, Any

def is_model_name(value: Any) -> TypeGuard[str]:
    return isinstance(value, str) and bool(value.strip())

try:
    resp = litellm.completion(model=model, messages=m)
except ValueError as e:
    if 'model param not passed in' in str(e):
        raise RuntimeError('model resolution produced None; check config/env keys') from e
    raise

When it happens

Trigger: Calling litellm.completion(model=my_model, ...) where my_model is None -- typically a config key typo, an os.environ.get that returned None, or a dict lookup for the model name that missed.

Common situations: MODEL env var unset in one environment; config-driven model selection with a missing key; refactoring that renamed the variable holding the model but left the call with the old (empty) one; default argument chains ending in None.

Understand the failure class

Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.

Related errors

• bucket_name must be provided for GCS destination
• max_budget cannot be negative. Received: {data.max_budget}
• soft_budget cannot be negative. Received: {data.soft_budget}
• soft_budget ({data.soft_budget}) must be strictly lower than
• Model '{m}' not in team's allowed models. Team allowed model

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/c2173f9ef35a40a0. Report an issue: GitHub.