* Existing usages of Serializer outside of this infrastructure should be migrated to use it.
* Existing usages of Java serialization should be migrated to use it.
* It is ok for Serializer to be used as a replacement for Java serialization as a migration step.

/// tip

You will normally have much better performance using a [Response Model](../tutorial/response-model.md) than returning a `JSONResponse` directly, as that way it serializes the data using Pydantic, in Rust.

///

## Return a `Response` { #return-a-response }

You can return a `Response` or any sub-class of it.

/// info

`JSONResponse` itself is a sub-class of `Response`.

///

        return ObjectInputFilter.Status.REJECTED;
    };

    /**
     * Tests if the object can be serialized.
     *
     * @param obj the object to be serialized (must not be {@literal null})
     * @return the deserialized object
     */
    public static Object serialize(final Object obj) {
        assertArgumentNotNull("obj", obj);

        final byte[] binary = fromObjectToBinary(obj);

    category=FastAPIDeprecationWarning,
    stacklevel=2,
)
class UJSONResponse(JSONResponse):
    """JSON response using the ujson library to serialize data to JSON.

    **Deprecated**: `UJSONResponse` is deprecated. FastAPI now serializes data
    directly to JSON bytes via Pydantic when a return type or response model is
    set, which is faster and doesn't need a custom response class.

    Read more in the

            raise ResponseValidationError(
                errors=errors,
                body=response_content,
                endpoint_ctx=ctx,
            )
        serializer = field.serialize_json if dump_json else field.serialize
        return serializer(
            value,
            include=include,
            exclude=exclude,
            by_alias=by_alias,
            exclude_unset=exclude_unset,

If you declare the return type as `AsyncIterable[Item]`, FastAPI will use it to **validate**, **document**, and **serialize** the data using Pydantic.

{* ../../docs_src/server_sent_events/tutorial001_py310.py ln[1:25] hl[10:12,23] *}

/// tip

As Pydantic will serialize it in the **Rust** side, you will get much higher **performance** than if you don't declare a return type.

///

If you declare the return type, FastAPI will use it to **validate** the data, **document** it in OpenAPI, **filter** it, and **serialize** it using Pydantic.

/// tip

As Pydantic will serialize it in the **Rust** side, you will get much higher **performance** than if you don't declare a return type.

///

### JSON Performance { #json-performance }

    ListenableFuture<@Nullable Void> unused =
        serializer.submitAsync(() -> settableFuture, directExecutor());
    for (int i = 0; i < 25_000; i++) {
      serializer.submit(Callables.<Void>returning(null), directExecutor()).cancel(true);
      unused = serializer.submit(Callables.returning(null), directExecutor());
    }
    ListenableFuture<Integer> stackDepthCheck =

                    not include that field, the JSON sent to the client would not have
                    that `password`.
                * Validation: whatever you return will be serialized with the
                    `response_model`, converting any data as necessary to generate the
                    corresponding JSON. But if the data in the object returned is not