unslothai / unsloth

Finetune Llama 3.2, Mistral, Phi, Qwen 2.5 & Gemma LLMs 2-5x faster with 80% less memory
https://unsloth.ai
Apache License 2.0
18.55k stars 1.3k forks source link

Bug fixes #1245

Closed danielhanchen closed 2 weeks ago