hrx: the kquant few-token kernels leave a weight prefill reads packed to the packed kernels (engine #2137) - #324

hrx: the kquant few-token kernels leave a weight prefill reads packed to the packed kernels (engine #2137)#324

bong-water-water-bong merged 2 commits into

1bit/hrx-vulkan-patched1bit-MONSTER/llama.cpp:1bit/hrx-vulkan-patchedfrom

cl-hrxmem/2137-few-token1bit-MONSTER/llama.cpp:cl-hrxmem/2137-few-tokenCopy head branch name to clipboard