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 #1682 from ZacharyZcR/fix/qwen-cuda-take-error
Vincenzo Fornaro62f6a47d9c5a
Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦