Clustering Nvidia DGX Spark and M3 Ultra Mac Studio for 4x Faster LLM Inferencetwitter.com8 points·alexandercheema··1 commentOpen articleSaveView on HN