llama.cpp
b0fb0f0a - ggml-cuda: fix blackwell native builds (#18361)

Commit
5 days ago
ggml-cuda: fix blackwell native builds (#18361) * ggml-cuda: fix blackwell native builds Replace 12x in native architectures by 12xa * replace for GGML_NATIVE=OFF too * only replace for native * remove 120f-virtual for default compilation --------- Co-authored-by: Aman Gupta <aman>
Author
Parents
Loading