llama.cpp
vulkan: add IQ3_S MMQ matmul kernels
#28822
Merged

vulkan: add IQ3_S MMQ matmul kernels #28822

vargad
vargad vulkan: add IQ3_S MMQ matmul kernels
3fe03cdf
vargad vargad requested a review 11 days ago
github-actions github-actions added Vulkan
github-actions github-actions added ggml
0cc4m
0cc4m commented on 2026-09-14
0cc4m
0cc4m commented on 2026-09-16
0cc4m
0cc4m approved these changes on 2026-09-16
vargad Make block_a_to_shmem do 2-byte loads (110 bytes is divisible by 2)
dd318442
vargad Align the check, IQ3_S is also using K tile size
486cbe1b
vargad vargad force pushed from 8165151f to 486cbe1b 7 days ago
0cc4m
0cc4m approved these changes on 2026-09-17
0cc4m 0cc4m added merge ready
ggerganov ggerganov merged bbd488c4 into master 6 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone