llama.cpp
Faster k-quants on older GPUs
#1930
Merged

Faster k-quants on older GPUs #1930

ggerganov merged 4 commits into master from ik/cuda-k-quants-2
ikawrakow
k_quants: hopefully much faster Q4_K on older GPUs
1677059b
k_quants: hopefully much faster Q3_K on older GPUs
be6f8b9e
k_quants: faster Q2_K on older GPUs
d6daebcb
k_quants: faster Q5_K on older GPUs
4aea4897
KerfuffleV2
JohannesGaessler
KerfuffleV2
johnson442
JohannesGaessler
ggerganov
ggerganov approved these changes on 2023-06-19
ggerganov ggerganov merged ca7c3f4d into master 2 years ago
ggerganov ggerganov deleted the ik/cuda-k-quants-2 branch 2 years ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone