langchain-ai/langgraph · error · ValueError

Invalid context schema type

Error message

Invalid context schema type: {context_schema}. Must be a BaseModel, TypedDict or dataclass.

What it means

Raised by get_context_jsonschema when the graph's context_schema is not a supported schema type. Only Pydantic BaseModel subclasses, TypedDicts, and dataclasses can be converted to JSON Schema; any other type annotation passed as context_schema fails here. Provide one of the supported schema types.

Solutions

1. Define context_schema as a pydantic BaseModel, TypedDict, or dataclass
2. Skip JSON-schema generation if an arbitrary type must be used

When it happens

Trigger: Thrown at libs/langgraph/langgraph/pregel/main.py:1003 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of langchain-ai/langgraph@38031739e5 (2026-08-26). Data as JSON: /api/errors/94c70857ad7974a2. Report an issue: GitHub.

Appendix: source

Thrown at libs/langgraph/langgraph/pregel/main.py:1003

            category=LangGraphDeprecatedSinceV10,
            stacklevel=2,
        )

        with warnings.catch_warnings():
            warnings.filterwarnings("ignore", category=LangGraphDeprecatedSinceV10)
            schema = self.config_schema(include=include)
        return schema.model_json_schema()

    def get_context_jsonschema(self) -> dict[str, Any] | None:
        if (context_schema := self.context_schema) is None:
            return None

        if isclass(context_schema) and issubclass(context_schema, BaseModel):
            return context_schema.model_json_schema()
        elif is_typeddict(context_schema) or is_dataclass(context_schema):
            return TypeAdapter(context_schema).json_schema()
        else:
            raise ValueError(
                f"Invalid context schema type: {context_schema}. Must be a BaseModel, TypedDict or dataclass."
            )

    @property
    def InputType(self) -> Any:
        if isinstance(self.input_channels, str):
            channel = self.channels[self.input_channels]
            if isinstance(channel, BaseChannel):
                return channel.UpdateType

    def get_input_schema(self, config: RunnableConfig | None = None) -> type[BaseModel]:
        config = merge_configs(self.config, config)
        if isinstance(self.input_channels, str):
            return super().get_input_schema(config)
        else:
            return create_model(
                self.get_name("Input"),
                field_definitions={

View on GitHub (pinned to 38031739e5)