Getting Started

Polyfactory is a simple and powerful mock data generation library, based around type hints and supporting dataclasses, TypedDicts, Pydantic models, msgspec Struct's and more.


pip install polyfactory


Minimal example using a dataclass
from dataclasses import dataclass

from polyfactory.factories import DataclassFactory

class Person:
    name: str
    age: float
    height: float
    weight: float

class PersonFactory(DataclassFactory[Person]):
    __model__ = Person

def test_is_person() -> None:
    person_instance =
    assert isinstance(person_instance, Person)
    assert isinstance(, str)
    assert isinstance(person_instance.age, float)
    assert isinstance(person_instance.height, float)
    assert isinstance(person_instance.weight, float)

That is it - with almost no work, we are able to create a mock data object fitting the Person class model definition.

This is possible because of the typing information available on the dataclass, which are used as a source of truth for data generation.

The factory parses the information stored in the dataclass and generates a dictionary of kwargs that are passed to the Person class constructor.

Relation to pydantic-factories

Prior to version 2, this library was known as pydantic-factories, a name under which it gained quite a bit of popularity.

A main motivator for the 2.0 release was that we wanted to support more than just Pydantic models, something which also required a change to its core architecture. As this library would no longer be directly tied to Pydantic, polyfactory was chosen as its new name to reflect its capabilities; It can generate mock data for dataclasses, TypedDicts, Pydantic, Odmantic, and Beanie ODM models, as well as custom factories.