Bunny v1.1 Llama 3 8b GGUF support/release?
Opened this issue · 1 comments
Deleted user commented
Hey all, thanks for your work. It seems there's an official gguf release for the 1.0 version of the Lllama-based model, but not of the 1.1 version. Is that because llama-cpp changes would be required? Or was interest just too low? I'm aware that one can quant one's own models, but lacking experience in this process, an official GGUF release would be quite nice.
Isaachhh commented
The source code of llama.cpp need to be modified to support S2-Wrapper, which is used in Bunny-v1.1.