ggml-cpu: unroll ggml_vec_dot_f16 into four accumulator groups to hide FMA latency#28

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

masterAvalancheHQ/llama.cpp:masterfrom

codspeed-optim-unroll-ggml-vec-dot-f16-into-four-accumulator-grou-1785215110005AvalancheHQ/llama.cpp:codspeed-optim-unroll-ggml-vec-dot-f16-into-four-accumulator-grou-1785215110005Copy head branch name to clipboard