Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
Merge pull request #1675 from ZacharyZcR/fix/cuda-mxfp4-scale-layout
Vincenzo Fornaro8ca90cd287f9
Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦