agentscope-ai/agentscope · error · ValueError

The structured_model is expected to be a subclass of Pydanti

Error message

The structured_model is expected to be a subclass of Pydantic.BaseModel or a dict, but got {type(structured_model)}.

What it means

_call_api_with_structured_output only accepts structured_model as a dict (JSON schema) or a Pydantic BaseModel subclass. Passing anything else (a class instance, a string, a TypedDict, a dataclass) raises ValueError.

Source

Thrown at src/agentscope/model/_base.py:721

                        _.input,
                        input_schema,
                    )
                    break

            if structured_output is None:
                raise StructuredOutputError(
                    "Failed to generate structured output for model.",
                )

            # Validate the output
            if isinstance(structured_model, dict):
                jsonschema.validate(structured_output, structured_model)

            elif issubclass(structured_model, BaseModel):
                structured_model.model_validate(structured_output)

            else:
                raise ValueError(
                    "The structured_model is expected to be a subclass of "
                    "Pydantic.BaseModel or a dict, "
                    f"but got {type(structured_model)}.",
                )
        except (
            ToolJSONDecodeError,
            jsonschema.ValidationError,
            PydanticValidationError,
        ) as e:
            raise StructuredOutputError(
                f"Invalid structured output from model {model_name}: {e}",
            ) from e

        return StructuredResponse(
            id=completed_response.id,
            created_at=completed_response.created_at,
            content=structured_output,
            usage=completed_response.usage,

View on GitHub (pinned to e90f1c7592)

Solutions

1. Pass the Pydantic class, not an instance: generate_structured_output(msgs, MySchema)
2. If you have a raw JSON schema, pass it as a dict: json.loads(schema_json)
3. Convert TypedDict/dataclass schemas to Pydantic BaseModel subclasses

Example fix

# before
res = await model.generate_structured_output(msgs, MySchema())

# after
res = await model.generate_structured_output(msgs, MySchema)

from pydantic import BaseModel
assert isinstance(structured_model, dict) or (isinstance(structured_model, type) and issubclass(structured_model, BaseModel))

from pydantic import BaseModel

def is_valid_schema(s) -> bool:
    return isinstance(s, dict) or (isinstance(s, type) and issubclass(s, BaseModel))

When it happens

Trigger: Calling generate_structured_output(msgs, MySchema()) with an instance instead of the class; passing a JSON string of a schema; passing TypedDict/dataclasses/attrs classes.

Common situations: Assuming an instantiated model works like OpenAI SDK's parse(); loading schema from JSON file as str; migrating code from pydantic v1 style.

Related errors

• Invalid structured output from model {model_name}: {e}
• The injection template must contain the '{runtime_state}' pl
• The input messages cannot be empty for the `generate_structu
• No structured-output strategy is available for {self.model}.
• Failed to get the completed response from model {model_name}

AI-assisted analysis of agentscope-ai/agentscope@e90f1c7592 (2026-08-28). Data as JSON: /api/errors/5d797ca4780e5638. Report an issue: GitHub.