llama.cpp

convert : make hf token optional

#14717

Merged