3 ms·
FWIW I run llama.cpp on AMD hardware using Vulkan. I've got no complaints but also nothing else to compare against.
by craftkiller 6mo ago
FWIW I run llama.cpp on AMD hardware using Vulkan. I've got no complaints but also nothing else to compare against.