Allow excluding defaults for specific values in serialization

Initial Checks

• I confirm that I'm using Pydantic V2

Description

The exclude_none option of model_dump and model_dump_json skips all fields that have the value of None. That works well if None is the default value of the field. However, the option also skips the field if it just has a value of None, but has no default value and is therefore required. This produces a ValidationError when trying to model_validate the output later.

This behavior is very surprising and unintuitive and makes the option very dangerous.

Would it be possible to narrow the scope of exclude_none only to the fields where None is the default? Essentially making it a subset of exclude_defaults that only excludes None defaults.

Example Code

from pydantic import BaseModel


class Foo(BaseModel):
    x: str | None


foo = Foo(x=None)
dumped = foo.model_dump_json(exclude_none=True)

Foo.model_validate_json(dumped)  # raises ValidationError: Field required

Python, Pydantic & OS Version

pydantic version: 2.13.4
        pydantic-core version: 2.46.4
          pydantic-core build: profile=release pgo=false
               python version: 3.14.3 (main, Mar 24 2026, 18:39:32) [Clang 17.0.0 (clang-1700.6.4.2)]
                     platform: macOS-26.6.2-arm64-arm-64bit-Mach-O
             related packages: mypy-2.3.1 typing_extensions-4.16.0
                       commit: unknown