llama.cpp
vulkan: In coopmat2 mmq, load q4_k/q5_k scales through shared memory
#12833

Merged

vulkan: In coopmat2 mmq, load q4_k/q5_k scales through shared memory #12833

0cc4m merged 1 commit into ggml-org:master from jeffbolznv:q45_k_scales

jeffbolznv requested a review from

0cc4m 341 days ago

github-actions added Vulkan

github-actions added ggml

vulkan: In coopmat2 mmq, load q4_k/q5_k scales through shared memory

8918306b

jeffbolznv force pushed to 8918306b 341 days ago

0cc4m approved these changes on 2025-04-09

0cc4m merged 0090950f into master 341 days ago

Reviewers

0cc4m

Assignees

No one assigned

Labels

Vulkan ggml

Milestone

No milestone