llama.cpp
metal: enable ggml-alloc
#2627
Merged

Loading