fastapi

FastAPI is a modern Python framework that builds async APIs with automatic OpenAPI/Swagger documentation. It combines Pydantic for data validation, async/await for non-blocking I/O, and dependency injection for modular service management. FastAPI generates interactive API docs automatically from your code.

Pydantic models define data schemas with automatic validation and serialization. In FastAPI, you declare Pydantic models as request bodies or response types, and FastAPI validates incoming data against those schemas before your endpoint runs, returning structured errors for invalid input.

Use FastAPI routers to organize endpoints into logical modules, then include them in your main app. Dependency injection patterns manage services and resources: declare dependencies as function parameters, and FastAPI resolves and injects them automatically, enabling testable, reusable code.

Yes. FastAPI runs on async frameworks like Uvicorn and supports async endpoint functions natively. Async endpoints handle multiple concurrent requests efficiently without blocking, ideal for database queries, HTTP calls, and WebSocket connections.

FastAPI includes built-in WebSocket support for bidirectional real-time communication. You define WebSocket routes alongside HTTP endpoints; FastAPI manages connection lifecycle and message handling, enabling live updates and interactive features without additional libraries.

Uvicorn is an async ASGI server ideal for development and single-process deployment. Gunicorn is a production WSGI server that manages multiple worker processes for horizontal scaling. Use Uvicorn for async-first deployments or Gunicorn with Uvicorn workers for multi-process production setups.