GitHub · C
benmaster82/picchio
A streaming Mixture-of-Experts (MoE) inference engine written in pure C, that runs models larger than your RAM on ordinary consumer hardware - MiniMax-M2 (230B) - GPT-OSS (20B and 120B) and Qwen3-MoE
- Stars
- 11
- Forks
- 3
- Stars gained in 24 hours
- 0
- Forks gained in 24 hours
- 0
Topics: inference-engine, inference-optimization, llm-inference, moe, avx2, cpu-inference, deep-learning, gpt-oss-120b, gpt-oss-20b, llm
Latest data point: Nov 21, 2026 · 11 stars · 3 forks