Nvidia DGX Spark and Apple Mac Studio = 4x Faster LLM Inference with EXO 1.0Comments on Hacker News | Source