Skip to content

cuda: fuse MoE weighted expert reduction - #25952

Merged
ORippler merged 3 commits into
ggml-org:masterfrom
anujj:cuda-moe-weighted-reduction-upstream
Sep 1, 2026
Merged

cuda: fuse MoE weighted expert reduction#25952
ORippler merged 3 commits into
ggml-org:masterfrom
anujj:cuda-moe-weighted-reduction-upstream