search

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
18.18k stars 1.27k forks source link

GGUF breaks #1212

Open awesomecoolraj opened 2 weeks ago

awesomecoolraj commented 2 weeks ago

This did not happen with older unsloth versions. What is happening here?

danielhanchen commented 2 weeks ago

Investigating!

danielhanchen commented 2 weeks ago

It should work fine @awesomecoolraj - is this on a Colab / Kaggle? Another option is to try https://github.com/unslothai/unsloth/wiki#manually-saving-to-gguf