frink

Pure-Rust GGUF inference engine with quantized CPU, Metal & CUDA kernels, MoE support, OpenAI-compatible server. Benchmarked head-to-head against llama.cpp.

What is frink?

Pure-Rust GGUF inference engine with quantized CPU, Metal & CUDA kernels, MoE support, OpenAI-compatible server. Benchmarked head-to-head against llama.cpp.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/antonelloffrink.svg)](https://yunodb.com/tools/antonelloffrink?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/antonelloffrink?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/antonelloffrink.svg" alt="Featured on YunoDB" height="20"></a>