TechEngineering at Meta1h ago
Meta details Prometheus, its gigawatt-scale supercluster in Ohio
Meta details Prometheus, its gigawatt-scale supercluster in Ohio, which will have ~1M GPUs at full scale and was used to train its Muse foundational models
TL;DRMeta built a massive Ohio data center with ~1M GPUs to train its new AI models.
Why it matters: Gigawatt-scale infrastructure reveals the hardware arms race defining AI leadership and computational costs.
- With the recent release of the Muse family of foundational models, we want to take you inside the infrastructure that has made it all possible.
Read full articleSource: Engineering at Meta · Opens in new tab