llama.cpp

convert.py : add consolidated.safetensors for mixtral 8x22b

#6587

Merged