mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-07-03 17:50:26 +00:00
d6e7b033a4
* llama : add option to save memory in device buffers * tests : extend llama-save-load-state