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 #1755 from namespaceMarcello/fix/qwen36-busy-slots
Vincenzo Fornaro94159eecd4a6
Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦