RT @alexocheema: Had to stack up 8 Mac Minis to get it running. ~5 tok/sec for now. First time running inference on 8 Mac Minis - perform…