llama.cpp
548e4e0d - ggml : check compute for fusion

Loading