Add sample JSON
Paste a JSON object or array, or upload a valid .json file up to 10 MB.
Python model generator
JSON to Python Class Converter
Turn sample JSON into structured Python dataclasses or Pydantic models with nested classes, accurate type hints, optional fields, aliases, and import-ready code.
JSON Input
Python Output
Local browser processing Nested type inference No signup
How to Convert JSON to Python Class
Generate strongly typed models from representative JSON in three simple steps.
Choose model options
Select dataclasses or Pydantic, set the root class name, and control optional, null, and naming behavior.
Export Python
Review the inferred definitions, copy them, or download a ready-to-edit .py file.
Python-Aware Inference Across the Full Sample
The converter examines every object in a top-level array, combines available keys, identifies fields missing from some records, and produces unions where values use different Python types.
Nested objects become named classes, object arrays receive typed item models, empty arrays use list[Any], and null values can become explicit None unions. Invalid keys are converted into safe attributes, with Pydantic aliases preserving the original JSON names.
- Infers str, int, float, bool, None, objects, lists, and mixed unions
- Generates import-ready dataclasses or Pydantic BaseModel classes
- Creates valid PascalCase classes and optional snake_case attributes
JSON
Python
Why Use This JSON to Python Class Converter?
Designed for API responses, SDK models, frontend data contracts, fixtures, migrations, and rapid prototyping.
Private generation
Your JSON is analyzed locally and is never submitted to the website server.
Array-aware models
All sampled array items contribute to property and union inference instead of only the first row.
Reviewable output
Generated classes include required imports and use Any only when the sample cannot support a narrower type.
Frequently Asked Questions
Important details about inference and generated Python definitions.
Can it generate dataclasses and Pydantic models?
Yes. Select standard-library @dataclass output or Pydantic BaseModel classes before generating.
How are arrays of objects handled?
The converter examines all sampled objects, combines their keys, marks missing properties as optional, and creates a named item model.
What happens to null values?
When null preservation is enabled, JSON null becomes None in a Python union, such as str | None.
How are unusual JSON keys handled?
Keys are converted to valid Python attribute names. Pydantic output adds Field(alias=...) when the generated name differs.
Is my JSON uploaded to the server?
No. Parsing, inference, preview, copying, and download creation happen locally in your browser.
Should I review generated classes?
Yes. The classes describe the supplied sample, so compare them with the complete API schema and validation rules before production use.