llama.cpp
CUDA: use mul_mat_q kernels by default
#2683
Merged
Overview
Commits
1
Changes
View On GitHub
FAQ Terms Privacy Refunds Impressum
Loading