Quantized GGUF models - desired contribution?
#3
by
marksverdhei
- opened
I noticed that there are quantized versions of norallm/normistral-7b-warm-instruct, but none for this model.
This isn't an instruct model, so that could be why, but I'm doing quantizations of this model myself atm,
so I was wondering if it would be preferred to have it merged into this repository instead of my own.
Cheers
Hi Mark, thanks! This is awesome, I'm happy to merge the quantized weights!
@davda54 I submitted a PR a week ago. Did you see it? Worked a week ago, but seems that he HF website breaks now when entering it.