llama.cpp
vulkan: optimize decodeFuncB in coopmat2 mul_mat_id shader
#18349

Merged

vulkan: optimize decodeFuncB in coopmat2 mul_mat_id shader #18349

0cc4m merged 2 commits into ggml-org:master from jeffbolznv:opt_decodefuncb

vulkan: Use BK=32 for coopmat2 mul_mat_id

b38ac9a8

vulkan: optimize decodeFuncB in coopmat2 mul_mat_id shader

8a72d2f4

jeffbolznv requested a review from

0cc4m 20 days ago

github-actions added Vulkan

github-actions added ggml

0cc4m approved these changes on 2025-12-26

0cc4m merged 7ac89021 into master 18 days ago

Reviewers

0cc4m

Assignees

No one assigned

Labels

Vulkan ggml

Milestone

No milestone