]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 38355c6)
vulkan: Implement topk_moe fused shader, ported from CUDA (#16641)
authorJeff Bolz <redacted>
Sat, 18 Oct 2025 10:22:57 +0000 (05:22 -0500)
committerGitHub <redacted>
Sat, 18 Oct 2025 10:22:57 +0000 (12:22 +0200)
commite56abd2098dd2e2b0804691b93c13b48ae421627
tree0b672f335d2639d91c973b916a65ecb3b8c654e3tree
parent38355c6c8e43204e11a22daa7483082c0ff01e71commit | diff
vulkan: Implement topk_moe fused shader, ported from CUDA (#16641)

This is similar to the CUDA shader from #16130, but doesn't use shared memory
and handles different subgroup sizes.
ggml/src/ggml-impl.h diff | blob | history
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
ggml/src/ggml-vulkan/vulkan-shaders/topk_moe.comp [new file with mode: 0644] blob
ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp diff | blob | history
Packaging of ggml-org/llama.cpp