mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-09-07 21:41:08 +02:00
e52aba537a
This allows to reduce compile time when you are building for a single GPU.