37 tok/s from Qwen3-Next-80B (48 GB) on a 32 GB laptop, 24 without drafts vs 2.8 for llama.cpp; reading 62 vs 16 tok/s. Faster than llama.cpp on all 13 models compared in Q4_K_M. LLMs bigger than your RAM, pure C, CPU or GPU.
C avx2ccpu-inferenceggufinference-enginelinuxllmlocal-llmmixture-of-expertsmoenvmequantizationqwen3simdspeculative-decodingvulkan
Insights
Commit activity
90 commits in the last 52 weeks (default branch, merge commits not counted).
Code frequency
+548,703 additions and −11,103 deletions in the last 52 weeks.
Contributors 1
Everyone who committed to the default branch.
| Contributor | Commits | Additions | Deletions |
|---|---|---|---|
camauri | 90 | +548,703 | −11,103 |