llama.cpp

5d5cb4c3 - ggml-meta: propagate buffer usage and call init on the new tensors (#27586)