b1280
da040034 · ggml-cuda : perform cublas fp16 matrix multiplication as fp16 (#3370) ·