mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-10-30 23:57:56 +01:00
This allows to reduce compile time when you are building for a single GPU.
This allows to reduce compile time when you are building for a single GPU.