3 ms·Clustering Nvidia DGX Spark and M3 Ultra Mac Studio for 4x Faster LLM Inference5 points by alexandercheema 1y ago