GGUF
Not-For-All-Audiences
nsfw

Unquantized Files for this Massive Model

#1
by Joseph717171 - opened

I appreciate your thoughtful quantizations; however, I would like to download the unquantizied repository and quantize the model down using the new SOTA 2 Quants that are supported in llama.cpp (main). Could you please upload the unquantified model to your repo? If I had more RAM, I would just run one of your other MoEs. But, unfortunately, I don't so I would greatly appreciate it if you could upload it to your repo. Thanks in advance. πŸ™πŸ€”

Joseph717171 changed discussion title from Unquantified Files for this Massive Model to Unquantized Files for this Massive Model

Thank you β€οΈπŸš€

Joseph717171 changed discussion status to closed

Sign up or log in to comment