failspy

/

Meta-Llama-3-70B-Instruct-abliterated-v3.5-GGUF

like 17

Howdy,

Great model. I'd love to experiment deeper, any chance you could release a slightly smaller q8 version of the model? I have 3x 3090s and I'm just a hair to short on the VRAM to load it all up.

Thanks for working on this project.