llama.cpp
CUDA: faster FA for GQA > 1 but not power of 2
#19092
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
1
Changes
View On
GitHub
CUDA: faster FA for GQA > 1 but not power of 2
#19092
JohannesGaessler
merged 1 commit into
ggml-org:master
from
JohannesGaessler:cuda-fa-gqa20-4
CUDA: faster FA for GQA > 1 but not power of 2
66f7a86c
github-actions
added
Nvidia GPU
github-actions
added
python
github-actions
added
ggml
ggerganov
approved these changes on 2026-01-25
JohannesGaessler
merged
0c21677e
into master
28 days ago
Login to write a write a comment.
Login via GitHub
Reviewers
ggerganov
Assignees
No one assigned
Labels
Nvidia GPU
python
ggml
Milestone
No milestone
Login to write a write a comment.
Login via GitHub