mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-06-28 15:20:20 +00:00
d6e7b033a4
* llama : add option to save memory in device buffers * tests : extend llama-save-load-state