llama.cpp
ggml-webgpu: improve i-quants mul_mat performance and speed up prefill
#24530

Merged

ggml-webgpu: improve i-quants mul_mat performance and speed up prefill #24530

reeselevine merged 2 commits into ggml-org:master from yomaytk:improve-mul_mat-iquants

Improve prefill speeds for i-quants

5fb0c144

yomaytk requested a review 42 days ago

github-actions added ggml

github-actions added WebGPU

CISC commented on 2026-06-12

Fix #if defined() usage in preprocessor guards.

1060fdc3

CISC approved these changes on 2026-06-13

reeselevine approved these changes on 2026-06-15

reeselevine merged 6e9007ae into master 39 days ago

yomaytk deleted the improve-mul_mat-iquants branch 39 days ago

Reviewers

reeselevine

CISC

Assignees

No one assigned

Labels

ggml WebGPU

Milestone

No milestone