microsoft/autogen · error · ValueError

model is required for OllamaChatCompletionClient

Error message

model is required for OllamaChatCompletionClient

What it means

OllamaChatCompletionClient's constructor requires a 'model' keyword argument. Unlike some clients, there is no default or environment fallback: if kwargs contains no 'model' key, __init__ raises ValueError immediately. The model name is needed both for the API call and to look up default capabilities.

Source

Thrown at python/packages/autogen-ext/src/autogen_ext/models/ollama/_ollama_client.py:976

            response_format=StructuredOutput,
        )
        result = await ollama_client.create([UserMessage(content="Who was the first man on the moon?", source="user")])  # type: ignore
        print(result)

    Note:
        Tool usage in ollama is stricter than in its OpenAI counterparts. While OpenAI accepts a map of [str, Any], Ollama requires a map of [str, Property] where Property is a typed object containing ``type`` and ``description`` fields. Therefore, only the keys ``type`` and ``description`` will be converted from the properties blob in the tool schema.

    To view the full list of available configuration options, see the :py:class:`OllamaClientConfigurationConfigModel` class.

    """

    component_type = "model"
    component_config_schema = BaseOllamaClientConfigurationConfigModel
    component_provider_override = "autogen_ext.models.ollama.OllamaChatCompletionClient"

    def __init__(self, **kwargs: Unpack[BaseOllamaClientConfiguration]):
        if "model" not in kwargs:
            raise ValueError("model is required for OllamaChatCompletionClient")

        model_capabilities: Optional[ModelCapabilities] = None  # type: ignore
        copied_args = dict(kwargs).copy()
        if "model_capabilities" in kwargs:
            model_capabilities = kwargs["model_capabilities"]
            del copied_args["model_capabilities"]

        model_info: Optional[ModelInfo] = None
        if "model_info" in kwargs:
            model_info = kwargs["model_info"]
            del copied_args["model_info"]

        client = _ollama_client_from_config(copied_args)
        create_args = _create_args_from_config(copied_args)
        self._raw_config: Dict[str, Any] = copied_args
        super().__init__(
            client=client, create_args=create_args, model_capabilities=model_capabilities, model_info=model_info
        )

View on GitHub (pinned to 027ecf0a37)

Solutions

1. Pass model='...' as a keyword argument: OllamaChatCompletionClient(model='llama3.1')
2. If loading from config, validate the dict contains the 'model' key before constructing and fix the config source
3. Check for misspellings like model_name or model_name_or_path and rename to model

Example fix

# before
client = OllamaChatCompletionClient(host='http://localhost:11434')  # no model

# after
client = OllamaChatCompletionClient(model='llama3.1', host='http://localhost:11434')

config = {'host': 'http://localhost:11434'}
assert 'model' in config, 'Ollama client config must include the model name'
client = OllamaChatCompletionClient(**config)

try:
    client = OllamaChatCompletionClient(**config)
except ValueError as e:
    if 'model is required' in str(e):
        config = {**config, 'model': os.environ['OLLAMA_MODEL']}
        client = OllamaChatCompletionClient(**config)
    else:
        raise

When it happens

Trigger: Constructing OllamaChatCompletionClient() with no arguments, or passing the model under a different key (model_name=..., model_id=...) so 'model' is absent from kwargs. Also occurs when loading from a config/component dict that lacks the 'model' key.

Common situations: Loading client config from JSON/YAML where the model field was omitted or misspelled; component-based deserialization of a config blob saved without the model; migrating code where another client used a different parameter name.

Related errors

• Required create args are missing: {required_create_args - cr
• modelName is a required property for LMStudioConfig and cann
• Could not create chat manager; make sure that it contains a
• Expected Memory, List[Memory], or None, got {type(memory)}
• At least one participant is required.

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/95da2348c246ad62. Report an issue: GitHub.