MoE streaming: expert reads in 256 KiB chunks by 16 threads (pin llama.cpp-vulkan cecf3ee) - #167

MoE streaming: expert reads in 256 KiB chunks by 16 threads (pin llama.cpp-vulkan cecf3ee)#167

bong-water-water-bong merged 2 commits into

main1bit-MONSTER/engine:mainfrom

moe/chunked-reads1bit-MONSTER/engine:moe/chunked-readsCopy head branch name to clipboard