llama.cpp

5364ae4b - llama : print hint when loading a model when no backends are loaded (#13589)