Spaces:
Running on A10G
Error quantizing: b'/bin/sh: 1: ./llama.cpp/quantize: not found\n'
I get this error on every model.
I'm hitting the same problem.
Same here
see this change on llama.cpp: https://github.com/ggerganov/llama.cpp/pull/7809#issuecomment-2171807144
We'll fix today! cc @reach-vb
Still getting this error on this model: BirdL/DeepSeek-Coder-V2-Lite-Instruct-FlashAttnPatch
Sorry for the delay on this! - looking into this right now.
Patched it, it should be fixed now! 🤗
Could some of you try it on different model checkpoints and ping if it works for use-cases as well!
(I'll close the issue once some of you confirm)
Patched it, it should be fixed now! 🤗
Could some of you try it on different model checkpoints and ping if it works for use-cases as well!
(I'll close the issue once some of you confirm)
Just created this gguf: https://huggingface.co/NikolayKozloff/Llama-3-8B-Swedish-Norwegian-Danish-chekpoint-18833-1-epoch-15_6_2024-Q8_0-GGUF Thank you very much.
Seems to work again, thanks!
Closing this as fixed! 🤗