llama.cpp

[ggml-virtgpu] Add a GHA build check

#22943

Merged