mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-05-13 04:24:17 +00:00
879d673759
* vulkan: Implement top-k Each pass launches workgroups that each sort 2^N elements (where N is usually 7-10) and discards all but the top K. Repeat until only K are left. And there's a fast path when K==1 to just find the max value rather than sorting. * fix pipeline selection * vulkan: Add N-ary search algorithm for topk * microoptimizations
321 KiB
321 KiB