llama.cpp
CUDA/HIP: derive MMVQ nwarps from kernel launch bounds
#28339

Open

CUDA/HIP: derive MMVQ nwarps from kernel launch bounds #28339

ACNoonan wants to merge 1 commit into ggml-org:master from ACNoonan:fix-mmvq-generic-nwarps

ACNoonan requested a review from

IMbackK 4 hours ago

ACNoonan requested a review 4 hours ago

github-actions added ggml

github-actions added CUDA

CUDA: derive MMVQ nwarps from kernel launch bounds

61535a26

ACNoonan force pushed from 8dc06299 to 61535a26 2 hours ago

Reviewers

IMbackK

Assignees

No one assigned

Labels

ggml CUDA

Milestone

No milestone