A fast asynchronous HTTP server and framework written in Rust for Python. To live and die by speed.
This project is intermittently maintained i.e as and when I have some free time to do so. Generally, the LNX project and Lust project take precedence.
Probably not, sure Litmus benches considerably faster than uvicorn with your typical small, plain text response benchmarks (generally 70%+ throughput increase) as soon as you put this into a real world situation I doubt you're going to get over 10% increase, you will probably get lower average latency but again by margins of around 10% so pick and choose your poison.
The main server api has been implemented other than the H2
and WS
protocol sections.