Unsloth

Unsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.

aniruddhaadak80 Updated 0 repo stars

File contents

aniruddhaadak80/herclew/tree/main/core-agent/optional-skills/mlops/training/unsloth commit dcadded527

Frequently asked questions

npx skillmds@latest add aniruddhaadak80/unsloth