microsoft/autogen · error · ValueError

model is required for AnthropicChatCompletionClient

Error message

model is required for AnthropicChatCompletionClient

What it means

AnthropicChatCompletionClient.__init__ hard-requires the 'model' kwarg because every subsequent create call needs it and Anthropic has no default model. The first check in the constructor raises ValueError('model is required for AnthropicChatCompletionClient') when it is absent — before any network or credential work happens.

Source

Thrown at python/packages/autogen-ext/src/autogen_ext/models/anthropic/_anthropic_client.py:367

            ToolParam(
                name=tool_schema["name"],
                input_schema=tool_params,
                description=tool_schema.get("description", ""),
            )
        )

        # Check if the tool has a valid name
        assert_valid_name(tool_schema["name"])

    return result


def normalize_name(name: str) -> str:
    """

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

        self._raw_config: Dict[str, Any] = dict(kwargs).copy()
        copied_args = dict(kwargs).copy()

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

        client = _anthropic_client_from_config(copied_args)
        create_args = _create_args_from_config(copied_args)

        super().__init__(
            client=client,
            create_args=create_args,
            model_info=model_info,
        )

View on GitHub (pinned to 027ecf0a37)

Solutions

1. Pass model explicitly: AnthropicChatCompletionClient(model='claude-sonnet-4-5', api_key=...).
2. Fail fast at app startup: assert config.get('model') is truthy before building the client.
3. When loading from component config, verify the dumped dict retains 'model' (check exclude_none/exclusion behavior).

Example fix

# before
client = AnthropicChatCompletionClient(api_key=os.environ['ANTHROPIC_API_KEY'])  # ValueError

# after
client = AnthropicChatCompletionClient(model=os.environ['ANTHROPIC_MODEL'], api_key=os.environ['ANTHROPIC_API_KEY'])

model = os.environ.get('ANTHROPIC_MODEL')
if not model:
    raise RuntimeError('ANTHROPIC_MODEL is not set; cannot build Anthropic client')
client = AnthropicChatCompletionClient(model=model, api_key=os.environ['ANTHROPIC_API_KEY'])

When it happens

Trigger: AnthropicChatCompletionClient(api_key=...) with no model; constructing from a config dict where 'model' is missing or None (None also fails since only presence is checked via 'model' not in kwargs... presence with None passes init but fails later at the API); env-driven configs where ANTHROPIC_MODEL was never set so the key is omitted.

Common situations: Porting from OpenAI client where a base-URL default model sometimes works; config loading with .get() returning None or the key dropped by exclude_none; tutorials omitting model in one snippet.

Related errors

• Required create args are missing: {required_create_args - cr
• Disallowed create args are present: {disallowed_create_args.
• config is required when using local Mem0 client (is_cloud=Fa
• tool_choice must be a Tool object, 'auto', 'required', or 'n
• Unknown content type: {part}

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/6137ca32d01cdb71. Report an issue: GitHub.