How to convert to HF format?
#6
by
ddh0
- opened
Hi, how can one convert this model to the HuggingFace Transformers format? config.json and other usual files seem to be missing.
I think they did this on purpose to force the use of their proprietary mistral inference, which default enrols you and your data into some sort of telemetry signal.
I would make an AWQ if I could generate or find the missing configs.
You can find the HF format here https://huggingface.co/bullerwins/Codestral-22B-v0.1-hf (thanks @bullerwins for the quick convert!) and GGUFs here
You can find the HF format here https://huggingface.co/bullerwins/Codestral-22B-v0.1-hf (thanks @bullerwins for the quick convert!) and GGUFs here
Thank you!
I have also uploaded EXL2 quants here:
