llama.cpp
CUDA: fewer memory bank conflicts for mul_mat_q
#2458
Merged
Overview
Commits
1
Changes
View On GitHub

CUDA: fewer memory bank conflicts for mul_mat_q #2458

JohannesGaessler merged 1 commit into ggml-org:master from JohannesGaessler:cuda-faster-mmq-2
JohannesGaessler
JohannesGaessler CUDA: fewer memory bank conflicts for mul_mat_q
5073b0f5
JohannesGaessler JohannesGaessler force pushed to 5073b0f5 2 years ago
slaren
JohannesGaessler
slaren
slaren approved these changes on 2023-07-31
JohannesGaessler JohannesGaessler merged 2dbf5189 into master 2 years ago
whoreson
JohannesGaessler
whoreson
cebtenzzre
whoreson

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum