{"slug": "marin-535b-a23b-starts-training-in-the-open", "title": "Marin 535B-A23B Starts Training, in the Open", "summary": "Marin 535B-A23B, a 535B-parameter model with 23B active parameters, began training this week in a fully open process, according to Percy Liang's announcement on X. The run will use 18.75T tokens on 11 x GB200 NVL72 for about 3 months, totaling 2.7e24 FLOPs, with pretraining (80%) and midtraining (20%) phases. This is the largest public pretraining run to date, surpassing BigScience's BLOOM 176B model.", "body_md": "# Marin 535B-A23B Starts Training, in the Open\n\n🚢 Marin 535B-A23B started training this week! As usual, the whole process is open.\n\nVoyage plan: pretraining (80%) + midtraining (20%) on 18.75T tokens on 11 x GB200 NVL72 for ~3 months (2.7e24 FLOPs). Post-training will follow.\n\nBefore kicking off the run, we trained a 4-rung scaling ladder from 1.6B-A61M (48B tokens) to 27.7B-A1.2B (926B tokens) to debug issues, and to make a forecast of our hero run. This is by far our biggest run, so definitely expecting the unexpected.\n\n[Percy Liang](https://x.com/percyliang/status/2090918065634684997), announcing the run on X.\n\nThe Marin project is a great example of being open to AI. The whole process is public: the scaling ladder they ran to debug the pipeline before committing GPU-months to the real thing, the exact token counts, the exact FLOPs, even the admission that they’re “expecting the unexpected” on their biggest run yet.\n\nMost labs treat a run like this as a trade secret until the model ships. It’s great we see another public model build. The last public run at this scale was [BLOOM](https://huggingface.co/bigscience/bloom), BigScience’s 176B model. Marin’s 535B total parameters (23B active) puts it well past that, the biggest public pretraining run that I’m aware of.", "url": "https://wpnews.pro/news/marin-535b-a23b-starts-training-in-the-open", "canonical_source": "https://julin.ai/2026/08/24/marin-open-training/", "published_at": "2026-08-23 12:00:00+00:00", "updated_at": "2026-08-26 23:19:35.051608+00:00", "lang": "en", "topics": ["large-language-models", "ai-research", "ai-infrastructure"], "entities": ["Marin 535B-A23B", "Percy Liang", "BLOOM", "BigScience", "GB200 NVL72"], "alternates": {"html": "https://wpnews.pro/news/marin-535b-a23b-starts-training-in-the-open", "markdown": "https://wpnews.pro/news/marin-535b-a23b-starts-training-in-the-open.md", "text": "https://wpnews.pro/news/marin-535b-a23b-starts-training-in-the-open.txt", "jsonld": "https://wpnews.pro/news/marin-535b-a23b-starts-training-in-the-open.jsonld"}}