Meta Unveils Gigawatt Prometheus Supercluster, Targets One Million GPUs

Meta's engineering blog details Prometheus, its gigawatt-scale supercluster in New Albany, Ohio, which trained the Muse foundation models with over 85% goodput and is being scaled toward about one million interconnected GPUs spanning more than 100 kilometers.

2026-10-19 ~ 2026-10-20 · 4 related posts

• Meta details Prometheus gigawatt cluster: 100km interconnects, ~1M GPUs, 99% fewer training interruptions — ryanshrout · 2026-10-19
• Meta details Prometheus gigawatt supercluster: 85%+ pretraining goodput for Muse — AIatMeta · 2026-10-20

2 near-duplicate retellings: BenBajarin · BenBajarin