search

unslothai / unsloth

Finetune Llama 3, Mistral, Phi & Gemma LLMs 2-5x faster with 80% less memory
https://unsloth.ai
Apache License 2.0
12.16k stars 789 forks source link

There is no file 'convert-lora-to-ggml.py' in llama.cpp folder any more #692

Open lab-peng opened 6 days ago

lab-peng commented 6 days ago

'python convert-lora-to-ggml.py my-model' won't work because there is no file 'convert-lora-to-ggml.py' in llama.cpp folder any more.

danielhanchen commented 2 days ago

So sorry on the delay - just relocated to the US so am very slow.

Oh yep! llama.cpp moved that file to some other place - forget where