llama.cpp
cuda : add half2 __shfl_xor() for ROCm 5.5
#7263
Merged
Overview
Commits
1
Changes
View On GitHub
FAQ Terms Privacy Refunds Impressum
Loading