llama.cpp
vulkan: optimize matrix multiplication for narrow N
#31990
Merged

vulkan: optimize matrix multiplication for narrow N #31990

jeffbolznv
jeffbolznv jeffbolznv requested a review from ggerganov ggerganov 6 days ago
jeffbolznv jeffbolznv requested a review 6 days ago
github-actions github-actions added testing
github-actions github-actions added Vulkan
github-actions github-actions added ggml
0cc4m
jeffbolznv
0cc4m
jeffbolznv
jeffbolznv vulkan: optimize matrix multiplication for narrow N
66988b01
jeffbolznv Attempt to handle wave64
dd7ca69e
0cc4m add n=16 tile for cm1
2dc75973
0cc4m apply RDNA split_k heuristic to narrow_n generally
3b3e73d5
0cc4m
jeffbolznv jeffbolznv force pushed from adb45679 to 3b3e73d5 3 days ago
0cc4m
0cc4m approved these changes on 2026-11-13
0cc4m 0cc4m added merge ready
ngxson ngxson merged d20426bb into master 1 day ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone