wen 2 bit model version?
#5
by
froilo
- opened
wen 2 bit model version?
You can try this if you don't mind using a experimental version of llama cpp, I was able to get it running with 64GBs of RAM. However, I did have a little headroom to load 3 bits, but for some reason it wouldn't load likely a bug
https://huggingface.co./pmysl/c4ai-command-r-plus-GGUF