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> |
||
|---|---|---|
| .. | ||
| ggml-alloc.h | ||
| ggml-backend.h | ||
| ggml-metal.h | ||
| ggml.h | ||
| llama.h | ||