cturan/llama.cpp @ 3e0be1cacef290c99cbb99ceaa433b4344f87355

镜像来自 https://github.com/cturan/llama.cpp

Georgi Gerganov 654a67794f metal : add FA-vec kernel for head size 64 (#13583)		8 月之前
..
cmake	771d84371c scripts : update sync + fix cmake merge	9 月之前
include	10d2af0eaa llama/ggml: add LLM training support (#10544)	8 月之前
src	654a67794f metal : add FA-vec kernel for head size 64 (#13583)	8 月之前
.gitignore	17eb6aa8a9 vulkan : cmake integration (#8119)	1 年之前
CMakeLists.txt	9c404ed54c sycl: use oneDNN for matrices multiplication (#12972)	8 月之前