strix-halo-engine-lab

strix-halo-engine-lab

hyukudan

Exact LLM inference engine written from scratch for AMD Strix Halo (gfx1151, ROCm/HIP). Qwen3.8-Flash-Next on one node or two replicas; DeepSeek V4.1 Flash across two nodes over USB4 with an optional RTX PRO. Speculative decoding bit-identical to greedy, up to 255 tok/s on coding chat. MiniMax and GLM next.

38 Stars
6 Forks
0 Watchers
C++ Language
apache-2.0 License
100 SrcLog Score
Cost to Build
$2.76M
Market Value
$6.85M

Growth over time

3 data points  ·  2026-11-06 → 2026-11-20
Stars Forks Watchers
💬

How do you feel about this project?

Ask AI about strix-halo-engine-lab

Question copied to clipboard

What is the hyukudan/strix-halo-engine-lab GitHub project? Description: "Exact LLM inference engine written from scratch for AMD Strix Halo (gfx1151, ROCm/HIP). Qwen3.8-Flash-Next on one node or two replicas; DeepSeek V4.1 Flash across two nodes over USB4 with an optional RTX PRO. Speculative decoding bit-identical to greedy, up to 255 tok/s on coding chat. MiniMax and GLM next.". Written in C++. Explain what it does, its main use cases, key features, and who would benefit from using it.

Question is copied to clipboard — paste it after the AI opens.

🏆 Show off this rank

hyukudan/strix-halo-engine-lab on SrcLog (holo style)

Add it to a GitHub README or profile — the card updates as the ranking changes:

How to clone strix-halo-engine-lab

Clone via HTTPS

git clone https://github.com/hyukudan/strix-halo-engine-lab.git

Clone via SSH

[email protected]:hyukudan/strix-halo-engine-lab.git

Download ZIP

Download main.zip

Found an issue?

Report bugs or request features on the strix-halo-engine-lab issue tracker:

Open GitHub Issues