tiangolo/fastapi · error · PydanticV1NotSupportedError

pydantic.v1 models are no longer supported by FastAPI…

Error message

pydantic.v1 models are no longer supported by FastAPI. Please update the response model {type_!r}.

What it means

Raised by `create_model_field` (fastapi/utils.py:67) as `PydanticV1NotSupportedError` when `annotation_is_pydantic_v1(type_)` returns True for a response model (or field) type. FastAPI no longer supports `pydantic.v1` models as response models / field types; only Pydantic v2 models are accepted. The offending type is shown via `{type_!r}` so you can locate it.

Solutions

1. Migrate the model to Pydantic v2: `from pydantic import BaseModel` and update field syntax.
2. If you cannot migrate yet, wrap/convert the v1 model to a v2 model (or a plain type) before using it as a response model.
3. Disable response model generation for that endpoint with `response_model=None` if conversion is not feasible.

Example fix

// before
from pydantic.v1 import BaseModel
class Item(BaseModel):
    name: str

@app.get('/items', response_model=list[Item])
// after
from pydantic import BaseModel
class Item(BaseModel):
    name: str

@app.get('/items', response_model=list[Item])

import pydantic

def is_pydantic_v2(model) -> bool:
    bases = getattr(model, '__mro__', ())
    return any(b.__module__.startswith('pydantic') and b.__name__ == 'BaseModel'
               and not b.__module__.startswith('pydantic.v1') for b in bases)

# before using a model as response_model:
assert is_pydantic_v2(MyModel), 'migrate to pydantic v2'

def is_v2_model(type_: object) -> bool:
    import pydantic
    return isinstance(type_, type) and issubclass(type_, pydantic.BaseModel)

When it happens

Trigger: Declaring a path operation whose return type or `response_model` references a model imported from `pydantic.v1` (e.g. `from pydantic.v1 import BaseModel`). Using a third-party library that still ships pydantic v1 models as response types. Mixing `pydantic` and `pydantic.v1` BaseModels in the same app.

Common situations: Upgrading FastAPI to a version that dropped v1 support while still importing v1 models. Inheriting from `pydantic.v1.BaseModel` for backwards compatibility. A dependency returns a v1 model annotated as the response type.

Related errors

• validation error(s)
• pydantic.v1 models are no longer supported by FastAPI…
• {errors}
• {"errors": exc.errors(), "body": body.decode()}
• {"errors": exc.errors(), "body": body.decode()}

AI-assisted analysis of tiangolo/fastapi@3e8d1526d8 (2026-08-11). Data as JSON: /api/errors/f1c2c44a6cfbe32f. Report an issue: GitHub.

Appendix: source

Thrown at fastapi/utils.py:67

    "check that {type_} is a valid Pydantic field type. "
    "If you are using a return type annotation that is not a valid Pydantic "
    "field (e.g. Union[Response, dict, None]) you can disable generating the "
    "response model from the type annotation with the path operation decorator "
    "parameter response_model=None. Read more: "
    "https://fastapi.tiangolo.com/tutorial/response-model/"
)


def create_model_field(
    name: str,
    type_: Any,
    default: Any | None = Undefined,
    field_info: FieldInfo | None = None,
    alias: str | None = None,
    mode: Literal["validation", "serialization"] = "validation",
) -> ModelField:
    if annotation_is_pydantic_v1(type_):
        raise PydanticV1NotSupportedError(
            "pydantic.v1 models are no longer supported by FastAPI."
            f" Please update the response model {type_!r}."
        )
    field_info = field_info or FieldInfo(annotation=type_, default=default, alias=alias)
    try:
        return v2.ModelField(mode=mode, name=name, field_info=field_info)
    except PydanticSchemaGenerationError:
        raise fastapi.exceptions.FastAPIError(
            _invalid_args_message.format(type_=type_)
        ) from None


def generate_operation_id_for_path(
    *, name: str, path: str, method: str
) -> str:  # pragma: nocover
    warnings.warn(
        message="fastapi.utils.generate_operation_id_for_path() was deprecated, "
        "it is not used internally, and will be removed soon",

View on GitHub (pinned to 3e8d1526d8)