llama.cpp

ec7f3ac9 - llama : add support for Deepseek-R1-Qwen distill model (#11310)