Overview

Pydantic 2.0, released on July 1, 2023, is a complete rewrite with a Rust core (pydantic-core). Validations are 5-50x faster and the API has been modernized.

Main Features

Rust core (5-50x faster)

The validation engine is rewritten in Rust via pydantic-core, delivering massive performance gains on data validation and serialization.

python 
from pydantic import BaseModel
from datetime import datetime

class User(BaseModel):
    name: str
    email: str
    age: int
    registered_at: datetime

# Validation 5-50x faster than v1
u = User(
    name='Alice',
    email='alice@example.com',
    age=30,
    registered_at='2023-07-01T10:00:00',
)
print(u.model_dump())  # new v2 API

New API

The API is renamed for clarity: .dict() becomes .model_dump(), .json() becomes .model_dump_json(), and .parse_obj() becomes .model_validate().

python 
from pydantic import BaseModel, field_validator

class Product(BaseModel):
    name: str
    price: float
    stock: int = 0

    @field_validator('price')
    @classmethod
    def price_must_be_positive(cls, v):
        if v <= 0:
            raise ValueError('Price must be positive')
        return v

p = Product(name='Widget', price=9.99, stock=100)
print(p.model_dump_json())  # JSON serialization
print(p.model_json_schema())  # JSON schema

Custom serialization

The new serialization system allows customizing output with @field_serializer decorators and json/python modes.

python 
from pydantic import BaseModel, field_serializer
from datetime import datetime

class Event(BaseModel):
    title: str
    date: datetime

    @field_serializer('date')
    def serialize_date(self, v: datetime, _info):
        return v.strftime('%d/%m/%Y %H:%M')

e = Event(title='Python Conf', date='2023-07-01T14:00:00')
print(e.model_dump())
# {'title': 'Python Conf', 'date': '01/07/2023 14:00'}

Sources