Many people asked which multi-Mac mode to use in oMLX, so I made this image.
This is a AI post classified by AGK Intelligence as AI infra & evals (a tutorial), kept by the AI Radar because it carries real work, not commentary.
Many people asked which multi-Mac mode to use in oMLX, so I made this image. Two high-end Macs of the same class (e.g. 2x M5 Ultra) -> Use TP (Tensor Parallel). Each Mac runs half of every layer, so prefill and decode both get faster. Data goes back and forth on every layer, so you need a very fast link like TB5 with RDMA. Two smaller Macs that can't hold the model alone -> Use PP (Pipeline Parallel). The layers are split by memory so the model fits across both. It's mainly for running a model that doesn't fit on one Mac, and decode doesn't get faster. One big but slower Mac plus one small
Posted by Jun Kim (4.5k followers) 1 days ago · 131 likes · 10.5k views · view the original post on X. Kept by the AI Radar as AI infra & evals.
More AI work like this
- Enterprise AI is only as useful as the data and business context behind it. — @databricks
- DecodeWall: another benchmark I'm building, focused on concurrent calls. — @ivanfioravanti
- KindlingAI's GLM 5.3 Flash for one Spark recipe is here! 700k total KV on DGX OS (up to… — @mmastrac
- A small model coaching a giant? Aiona checks the held-out tie and SMF’s memory-card… — @MichaelGannotti
- Ranking inference providers six ways — @AdamHoltererer
- Harness adapters training is now live! — @yb2698
- NVIDIA is giving Kimi K3.1 Pro away for free. — @VaibhavSisinty
- The world’s first co-packaged optics Ethernet switch is in full production. — @NVIDIAAIInfra
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.6k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 16:05 UTC. Full method.