vulkan: add Intel Xe flash attention optimization kernels (2/3, Xe-LPG Plus/Xe2/Xe3) - #24406

vulkan: add Intel Xe flash attention optimization kernels (2/3, Xe-LPG Plus/Xe2/Xe3)#24406

fish-jiang wants to merge 1 commit into

ggml-org:masterggml-org/llama.cpp:masterfrom

fish-jiang:intel/xe-flash-attnfish-jiang/llama.cpp:intel/xe-flash-attnCopy head branch name to clipboard