llama.cpp
llama.cpp copied to clipboard
Add support for missing quants in CPY (Metal & CUDA).
Fixes #10976.