Skip to content

🐍 Generate from Python Models

Generate code from existing Python types: Pydantic models, dataclasses, TypedDict, or dict schemas. This is useful for converting between model types or generating from programmatically-defined schemas.

🚀 Quick Start

datamodel-codegen --input-model ./mymodule.py:User --output model.py

Format

--input-model <path/to/file.py>:<ObjectName>

Simply specify the file path and object name separated by :.

Example

myproject/
├── src/
│   └── models/
│       └── user.py      # class User(BaseModel): ...
└── schemas.py           # USER_SCHEMA = {...}

# From file path (recommended - easy to copy-paste)
datamodel-codegen --input-model src/models/user.py:User --output model.py
datamodel-codegen --input-model ./schemas.py:USER_SCHEMA --input-file-type jsonschema

# Windows paths also work
datamodel-codegen --input-model src\models\user.py:User

Copy-paste friendly

Just copy the file path from your editor or file explorer, add :ClassName, and you're done!

Module format (alternative)

You can also use Python module notation with dots:

datamodel-codegen --input-model src.models.user:User
datamodel-codegen --input-model schemas:USER_SCHEMA --input-file-type jsonschema

Current directory is auto-added to sys.path

No PYTHONPATH configuration needed.

File and package name conflict

If both mymodule.py and mymodule/ directory exist, use ./ prefix: 

datamodel-codegen --input-model ./mymodule.py:Model

Supported Input Types

Type Description Requires
Pydantic BaseModel Pydantic v2 models with model_json_schema() Pydantic v2
dataclass Standard library @dataclass Pydantic v2 (for TypeAdapter)
Pydantic dataclass @pydantic.dataclasses.dataclass Pydantic v2
TypedDict typing.TypedDict subclasses Pydantic v2 (for TypeAdapter)
dict Dict containing JSON Schema or OpenAPI spec -

Pydantic v2 Required

All Python type inputs (except raw dict) require Pydantic v2 runtime to convert to JSON Schema.

📝 Examples

Pydantic BaseModel

mymodule.py 

from pydantic import BaseModel

class User(BaseModel):
    name: str
    age: int


datamodel-codegen --input-model ./mymodule.py:User --output model.py

✨ Generated model.py 

from __future__ import annotations

from pydantic import BaseModel


class User(BaseModel):
    name: str
    age: int

Convert Pydantic to TypedDict

datamodel-codegen --input-model ./mymodule.py:User --output-model-type typing.TypedDict --output model.py

✨ Generated model.py 

from __future__ import annotations

from typing import TypedDict


class User(TypedDict):
    name: str
    age: int

Standard dataclass

mymodule.py 

from dataclasses import dataclass

@dataclass
class User:
    name: str
    age: int


datamodel-codegen --input-model ./mymodule.py:User --output model.py

TypedDict

mymodule.py 

from typing import TypedDict

class User(TypedDict):
    name: str
    age: int


datamodel-codegen --input-model ./mymodule.py:User --output model.py

Dict Schema (JSON Schema)

mymodule.py 

USER_SCHEMA = {
    "type": "object",
    "properties": {
        "name": {"type": "string"},
        "age": {"type": "integer"},
    },
    "required": ["name", "age"],
}


datamodel-codegen --input-model ./mymodule.py:USER_SCHEMA --input-file-type jsonschema --output model.py

Dict requires --input-file-type

When using a dict schema, you must specify --input-file-type (e.g., jsonschema, openapi).

Dict Schema (OpenAPI)

mymodule.py 

OPENAPI_SPEC = {
    "openapi": "3.0.0",
    "info": {"title": "API", "version": "1.0.0"},
    "paths": {},
    "components": {
        "schemas": {
            "User": {
                "type": "object",
                "properties": {
                    "name": {"type": "string"},
                    "age": {"type": "integer"},
                },
            }
        }
    },
}


datamodel-codegen --input-model ./mymodule.py:OPENAPI_SPEC --input-file-type openapi --output model.py

Mutual Exclusion

--input-model cannot be used with:

  • --input (file input)
  • --url (URL input)
  • --watch (file watching)

📖 See Also