You can now run Kimi K2.7 Code locally! π
We shrank the 1T model to 325GB (-48%) via Dynamic 2-bit where important layers are upcasted.
Run at >40 tok/s on 330GB RAM/VRAM setups.
Run full precision on 610 GB.
Guide: unsloth.ai/docs/models/kimi-β¦
GGUF: huggingface.co/unsloth/Kimi-β¦