Kaden-Schutt/hipfire
A native LLM inference engine for AMD RDNA GPUs built in Rust, shipping as a single binary with HIP kernels.
Not currently ranked — collecting fresh signals.
star history
Hipfire is an RDNA-native inference engine targeting the entire AMD RDNA family from RDNA1 through RDNA4 on consumer, pro, and APU hardware. It runs LLM inference without Python or the full ROCm userspace stack, instead shipping pre-compiled kernels and JIT-compiling the rest through HIP. Supports quantization, speculative decoding via DFlash, and provides an Ollama-style CLI with an OpenAI-compatible API daemon.
Frequently asked
- What is Kaden-Schutt/hipfire?
- A native LLM inference engine for AMD RDNA GPUs built in Rust, shipping as a single binary with HIP kernels.
- Is hipfire open source?
- Yes — Kaden-Schutt/hipfire is an open-source project tracked on heatdrop.
- What language is hipfire written in?
- Kaden-Schutt/hipfire is primarily written in Rust.
- How popular is hipfire?
- Kaden-Schutt/hipfire has 444 stars on GitHub.
- Where can I find hipfire?
- Kaden-Schutt/hipfire is on GitHub at https://github.com/Kaden-Schutt/hipfire.