llama.cpp
opencl: fold gqa heads for f16 decomposed FA
#32631
Merged

opencl: fold gqa heads for f16 decomposed FA #32631

shaofeiqi
github-actions github-actions added ggml
github-actions github-actions added OpenCL
lhez lhez changed the title opencl: fold gqa heads for f16 FA opencl: fold gqa heads for f16 FA for more efficient spec verify 9 hours ago
lhez lhez changed the title opencl: fold gqa heads for f16 FA for more efficient spec verify opencl: fold gqa heads for f16 decomposed FA 7 hours ago
lhez opencl: fold GQA head for f16 FA for better efficiency
1e09da99
lhez lhez force pushed from 5caaba0e to 1e09da99 7 hours ago
lhez lhez marked this pull request as ready for review 7 hours ago
lhez lhez requested a review 7 hours ago
lhez
lhez
lhez approved these changes on 2026-11-20
max-krasnyansky
max-krasnyansky approved these changes on 2026-11-20
lhez lhez merged 7e960f40 into master 5 hours ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone