1Kimi K2.5 on 2 Mac Studios (<$20K) running at 24 TPS (opens in new tab)(twitter.com)6alexandercheema5mo ago4Save
2Clustering Nvidia DGX Spark and M3 Ultra Mac Studio for 4x Faster LLM Inference (opens in new tab)(twitter.com)8alexandercheema8mo ago1Save
3Clustering Nvidia DGX Spark and M3 Ultra Mac Studio for 4x Faster LLM Inference (opens in new tab)(blog.exolabs.net)5alexandercheema8mo ago0Save
4Meta AI: "The Future of AI Is Open Source and Decentralized" (opens in new tab)(twitter.com)214alexandercheema1y ago111Save
5Two MacBooks is All You Need: Running Llama 3 405B using Apple MLX and Exo (opens in new tab)(twitter.com)11alexandercheema1y ago2Save
