llama.cpp
cuda: Support Q2_0
#25603
Open

cuda: Support Q2_0 #25603

dfriehs wants to merge 4 commits into ggml-org:master from dfriehs:q2_0-cuda
dfriehs
khosravipasha cuda: Q2_0
dd5decdb
dfriehs tests: enable Q2_0 for test-backend-ops perf
b35bf1c7
dfriehs cuda: adjust Q2_0 to new group size of 64
71bbe0bc
dfriehs cuda: extract Q2_0 elements via __byte_perm
d19075f4
dfriehs dfriehs requested a review from ggerganov ggerganov 1 day ago
dfriehs dfriehs requested a review 1 day ago
github-actions github-actions added testing
github-actions github-actions added ggml
github-actions github-actions added CUDA
Green-Sky
khosravipasha
JohannesGaessler
dfriehs dfriehs marked this pull request as draft 17 hours ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone