4bit?
https://huggingface.co/jondurbin/airoboros-13b-ggml-q4_0
Or the 7b version: https://huggingface.co/jondurbin/airoboros-7b-ggml-q4_0
ggml? but that's not gpu accelerated is it? I just looking for a the 4bit one that does work on gpu
· Sign up or log in to comment