Pydantic Model Generator

Pydantic is a Python data validation library that uses type annotations to define data schemas. When you create a model, Pydantic automatically validates input data, converts types (e.g. '123' to int), and raises descriptive errors for invalid data. It's the foundation of FastAPI's request/response handling and is widely used for config management, data parsing, and API clients.

Pydantic v2 (released 2023) was rewritten in Rust via pydantic-core, making it 5-50x faster than v1. Key API changes: validators now use @field_validator and @model_validator decorators (replacing @validator and @root_validator); orm_mode is now model_config = ConfigDict(from_attributes=True); class Config is replaced by model_config. Many v1 APIs still work via compatibility shim but are deprecated.

A field without a default is required. Use Optional[T] with a default of None for nullable optional fields: name: Optional[str] = None. Use a default value for non-nullable optionals: count: int = 0. Pydantic v2 uses from __future__ import annotations or explicit Optional to handle None. The Field() function provides richer defaults: field: str = Field(default='hello', min_length=1).

model_config replaces the inner Config class from v1. It accepts a ConfigDict with options like: from_attributes=True (replaces orm_mode, allows model creation from ORM objects), str_strip_whitespace=True (auto-strip strings), validate_default=True (validate default values), populate_by_name=True (allow both alias and field name). Example: model_config = ConfigDict(from_attributes=True, str_strip_whitespace=True).

@field_validator('field_name') validates a single field. @model_validator(mode='after') runs after all fields are validated and has access to the full model. Validators must be classmethods decorated with @classmethod. For simple value constraints, use Field(gt=0, le=100) instead of writing a validator.

model.model_dump() is the v2 method (replaces v1's .dict()). It returns a dictionary representation of the model. You can exclude fields: model.model_dump(exclude={'password'}), include only certain fields: model.model_dump(include={'id', 'name'}), or serialize aliases: model.model_dump(by_alias=True). model.model_dump_json() returns a JSON string directly.