PR #16754 CUDA: use CUB for arbitary size argsort

CUDA: use CUB for arbitary size argsort #16754

am17an merged 1 commit into ggml-org:master from am17an:cuda_cub_argsort

CUDA: use CUB for arbitary size argsort

fa43f8a3

am17an requested a review from

slaren 245 days ago

slaren commented on 2025-10-24

github-actions added Nvidia GPU

github-actions added ggml

slaren approved these changes on 2025-10-24

am17an merged 0bcb40b4 into master 245 days ago

am17an deleted the cuda_cub_argsort branch 245 days ago

Reviewers

slaren

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone