vibora-io/vibora

★ 5,598PythonAudience · developerComplexity · 3/5LicenseSetup · easy

Mindmap

mindmap
  root((vibora))
    Server side
      Async endpoints
      Sessions
      Websockets
      Streaming responses
    Client side
      Auto retries
      Rate limiting
      Connection pooling
      Cookie persistence
    Features
      Template rendering
      Built-in testing
      Type hints
    Status
      Alpha
      Active rewrite

mindmap root((vibora)) Server side Async endpoints Sessions Websockets Streaming responses Client side Auto retries Rate limiting Connection pooling Cookie persistence Features Template rendering Built-in testing Type hints Status Alpha Active rewrite

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Build a high-throughput async Python API server for a prototype or performance experiment.

USE CASE 2

Make HTTP client requests with automatic retries and connection pooling inside a Python service.

USE CASE 3

Experiment with Python async web performance benchmarks compared to Flask or Django.

Tech stack

Python

Getting it running

Difficulty · easy Time to first run · 30min

Project is in alpha and actively being rewritten, treat as experimental only, not suitable for production.

Use freely for any purpose including commercial projects, as long as you keep the copyright notice.

In plain English

Vibora is a Python web framework built for speed and asynchronous operation. It lets developers write both the server side (handling incoming web requests) and the client side (making outgoing HTTP requests) using the same library. The name means "viper" in Portuguese. On the server side, Vibora supports features like sessions, websockets, streaming responses, caching, and template rendering that updates automatically when files change. It also includes a built-in testing framework and leans on Python's type hint system to make code easier to read and validate. The basic server example in the README shows a working web endpoint in about eight lines of code. On the client side, the library can make web requests with automatic retries, rate limiting, connection pooling, and cookie persistence. It handles different authentication styles and transparently decompresses compressed responses. The headline claim is raw throughput. A benchmark table in the README shows Vibora handling around 368,000 requests per second on a simple test, compared to much lower numbers for Flask, Django, and other common Python frameworks. These numbers come from a "hello world" test, which is a best-case scenario and not representative of real application workloads. Importantly, the README opens with a warning that the project is being completely rewritten. At the time of writing the project was in alpha and the rewrite status is unclear. This means the library should be treated as experimental rather than production-ready. The MIT license allows free use, and documentation was available at a dedicated docs site.

Copy-paste prompts

Prompt 1

Show me the minimal Vibora code to create an async HTTP endpoint that returns a JSON response.

Prompt 2

How does Vibora's built-in HTTP client handle automatic retries and rate limiting when calling an external API?

Prompt 3

I want to stream a large response from a Vibora server to a client without loading it all into memory. How do I do that?

Prompt 4

Compare Vibora's async session and cookie handling to how you would do the same thing in aiohttp or httpx.

Open on GitHub → Explain another repo

← vibora-io on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.