llama.cpp

server : don't list cached models when a preset is used

#25226

Open