llama.cpp
CUDA: fastdiv, launch bounds for mmvq + q8_1 quant
#15802

Merged

CUDA: fastdiv, launch bounds for mmvq + q8_1 quant #15802

JohannesGaessler merged 2 commits into ggml-org:master from JohannesGaessler:cuda-mmvq-fastdiv-3

CUDA: fastdiv, launch bounds for mmvq + q8_1 quant

dbde65ff

github-actions added Nvidia GPU

github-actions added ggml

ORippler commented on 2025-09-05

use fastfdiv for mul_mat_id modulo

748c6a53

ORippler approved these changes on 2025-09-05

slaren approved these changes on 2025-09-05

JohannesGaessler merged 5143fa89 into master 62 days ago

Reviewers

slaren

ORippler

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone