Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
Merge #168: int3-g64 (fmt=5) per-group-scale 3-bit expert quant
Vincenzo Fornarof42adfae76e1
Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦