ggml-cpu: dequantize each weight row once per matmul tile (4-column Q4_0/Q4_K AVX2 kernels) - #41

ggml-cpu: dequantize each weight row once per matmul tile (4-column Q4_0/Q4_K AVX2 kernels)#41

codspeed-hq[bot] wants to merge 1 commit into

masterAvalancheHQ/llama.cpp:masterfrom

codspeed-optim-dequantize-each-weight-row-once-per-matmul-tile-4-1785338847703AvalancheHQ/llama.cpp:codspeed-optim-dequantize-each-weight-row-once-per-matmul-tile-4-1785338847703Copy head branch name to clipboard