General Availability Status
Python Workers have reached general availability, establishing Python as a first-class, fully supported language on the Cloudflare Developer Platform. This status allows developers to utilize existing Python code, libraries, and design patterns while connecting them seamlessly to platform services such as Workers AI, R2, D1, Hyperdrive, Durable Objects, Queues, and Workflows.
Performance Metrics
Cloudflare announced these advancements on December 8, 2025, highlighting that Python Workers achieve cold starts of approximately 1.027 seconds for packages like FastAPI. This performance is 2.4x faster than AWS Lambda without SnapStart, which records a 2.502-second cold start, and 3x faster than Google Cloud Run at 3.069 seconds.
Integration Mechanisms
Developers can run applications using the workers.asgi package by adding a specific snippet to their code, or utilize workers.wsgi for synchronous web applications such as Django. Additionally, Workers RPC now functions across Python and JavaScript in both directions without requiring protocol definitions or explicit serialization code at the boundary. Despite general availability, Python Workers inherit platform constraints because they run CPython compiled to WebAssembly through Pyodide. Consequently, performance-sensitive Python code remains bounded by the WASM interpreter rather than native execution.
Cost and Availability
Cloudflare offers Python Workers performance gains on its free tier, whereas AWS charges extra for SnapStart to achieve similar cold start improvements. The 2.4x cold start advantage is most applicable when comparing against Lambda configurations that do not utilize the paid SnapStart feature.
Native Binding Support
Python Workers now natively support Cloudflare Developer Platform bindings, removing the need for explicit conversion of Python objects into TypeScript objects at the RPC boundary. This native integration allows developers to connect existing Python code and libraries directly to platform services without manual serialization overhead. Developers building applications with relational databases such as PostgreSQL or MySQL can now integrate Hyperdrive into Python Workers. This capability extends the platform's database connectivity options to Python-based edge applications, supporting standard relational data workflows.
ASGI and WSGI Entry Points
Developers can deploy existing Python applications by integrating the workers.asgi package, which requires adding a specific snippet to define a WorkerEntrypoint. This approach uses an async fetch method that delegates processing to the asgi.fetch function, passing the application instance, request object, and environment variables. For synchronous web frameworks like Django, the platform provides the workers.wsgi package. Developers can instantiate a Default entry point by calling wsgi.entrypoint(app) after importing the application from their specific Django module structure.
Under-the-Hood Execution Model
The execution of Python Workers relies on CPython compiled to WebAssembly through Pyodide, which defines the technical foundation for running these applications. This architecture means that performance-sensitive code is bounded by the WASM interpreter rather than native machine execution. While this setup enables general availability and first-class language support, it introduces specific constraints inherent to the platform's runtime environment.
