llama.cpp
CUDA graphs on Pascal (sm_61): +40% on MoE, +7% on dense, no regression
#27721

Open

CUDA graphs on Pascal (sm_61): +40% on MoE, +7% on dense, no regression #27721

ArtyomITA wants to merge 1 commit into ggml-org:master from ArtyomITA:cuda-graphs-pascal-sm61

CUDA: enable CUDA graphs on Pascal (sm_61)

0cc7efed

ArtyomITA requested a review 1 day ago

github-actions added ggml

github-actions added CUDA

ggml-gh-bot added draft

github-actions marked this pull request as draft 1 day ago

github-actions removed draft

ArtyomITA marked this pull request as ready for review 1 day ago

Reviewers

No reviews

Assignees

No one assigned

Labels

ggml CUDA

Milestone

No milestone