0ceeb11721
* Merging mainline - WIP * Merging mainline - WIP AVX2 and CUDA appear to work. CUDA performance seems slightly (~1-2%) lower as it is so often the case with llama.cpp/ggml after some "improvements" have been made. * Merging mainline - fix Metal * Remove check --------- Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com> |
||
|---|---|---|
| .. | ||
| arm64-windows-llvm.cmake | ||
| arm64-windows-msvc.cmake | ||
| build-info.cmake | ||
| git-vars.cmake | ||
| llama-config.cmake.in | ||
| llama.pc.in | ||