llama.cpp
CUDA: tune GLM 4.7 Flash FA kernel selection logic
#19097

Merged

CUDA: tune GLM 4.7 Flash FA kernel selection logic #19097

JohannesGaessler merged 1 commit into ggml-org:master from JohannesGaessler:cuda-fa-gqa20-5

github-actions added Nvidia GPU

github-actions added ggml

ggerganov approved these changes on 2026-01-26

CUDA: tune GLM 4.7 Flash FA kernel selection logic

bee6c679

JohannesGaessler force pushed from 134adcac to bee6c679 132 days ago

JohannesGaessler merged a5bb8ba4 into master 131 days ago

Reviewers

ggerganov

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone