Ferrox

Pure-Rust GGUF inference engine with CPU, Metal and CUDA kernels, benchmarked vs llama.cpp.

PlatformWebfreeglobal

Ferrox is a pure-Rust GGUF inference engine supporting dense and Mixture-of-Experts models on CPU, Apple Metal or CUDA, with an OpenAI-compatible server, benchmarked head-to-head against llama.cpp. It targets developers running local LLM inference who want a native Rust engine rather than binding to a C++ library.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of Ferrox - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim Ferrox to unlock →

Something wrong with this listing — dead link, not a real product, wrong info?

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether Ferrox (or something else) actually fits.

Try a problem