Upload in splits of max 50GB due to HF 50GB limit. (made with llama.cpp commit a1d6df1)

GGUF model commit (made with llama.cpp commit a1d6df1)

initial commit