---
title: "Meta details Prometheus, its Ohio AI supercluster heading for about 1 million GPUs"
url: https://anawords.com/story/meta-prometheus-supercluster-ohio-1-million-gpus
published: 2026-10-20T13:37:16+05:30
updated: 2026-10-20T13:37:16+05:30
section: Chips & Compute
author: Mayank Sahu
checked: sources checked by AI, approved by editor
version: 1
---

# Meta details Prometheus, its Ohio AI supercluster heading for about 1 million GPUs

Meta’s engineers say the New Albany site trained its Muse models, now runs hundreds of thousands of GPUs, and cut training interruptions by 99% from a 2025 peak.

**In short:** Meta published a detailed look at Prometheus, its gigawatt-scale AI supercluster in New Albany, Ohio, on 19 October 2026. Meta says it runs hundreds of thousands of GPUs today, is being scaled toward roughly 1 million, and trained its Muse family of models.

## What Meta shared

Four Meta engineers described Prometheus in a post on Engineering at Meta.[^1] About two years ago, the New Albany site ran ordinary web traffic; it is now a “gigawatt-scale supercluster”.[^1] Meta does not give an exact power figure.[^1]

It has “hundreds of thousands of GPUs” and is being scaled toward “roughly 1 million”.[^1][^2] Its first phase trained the Muse family of models, including Muse Wonder.[^1][^2]

## The hardware

- **Catalina:** Nvidia GB200 NVL72, about 120kW per two-rack pair.[^1][^2]
- **Clemente:** Nvidia GB300 NVL72, 288GB of HBM3e per GPU, about 135kW per rack.[^1][^2]
- **San Miguel:** Nvidia Vera Rubin NVL72, coming soon, over 170kW per rack.[^1][^2]

Each building’s network connects up to 20,736 GPUs over Ethernet, with 800G of bandwidth per GPU, and a regional network links buildings from 2km to over 100km apart.[^1] Cooling mixes air-assisted liquid cooling, facility liquid cooling and closed-loop systems.[^1]

## Reliability

Meta reports more than 85% goodput on Muse Wonder pre-training and more than 90% on reinforcement learning runs.[^1][^2] Interruptions fell 99% from a mid-2025 peak of nearly 1,000 per 8,000 GPUs per day, and mean time between interruptions rose past 54,000 GPU hours, with a target of about 100,000.[^1][^2] These are Meta’s own measurements.

## What comes next

Lessons from Prometheus feed into Hyperion, Meta’s planned supercluster in Richland Parish, Louisiana, including closed-loop liquid cooling and a grid connection with Entergy.[^1]

## Why it matters

Few companies publish this level of detail. The reliability figures show how much engineering goes into keeping very large training runs going, beyond buying chips.

## Fact ledger

- [Confirmed] Prometheus is in New Albany, Ohio, and trained Meta’s Muse models. (source 1)
- [Confirmed] It has hundreds of thousands of GPUs and is being scaled toward roughly 1 million. (source 1)
- [Confirmed] It uses Nvidia GB200 and GB300 NVL72 racks, with Vera Rubin NVL72 coming. (source 1)
- [Company claim] Meta reports goodput above 85% for pre-training and above 90% for RL runs. (source 1,2)
- [Company claim] Interruptions fell 99% from a mid-2025 peak, per Meta. (source 1,2)
- [Not known] The site’s exact power capacity.

## Sources

1. [Engineering at Meta: Inside Prometheus, Meta’s journey to a gigawatt-scale supercluster](https://engineering.fb.com/2026/10/19/data-center-engineering/inside-prometheus-meta-gigawatt-scale-supercluster/) — official
2. [StorageReview: Meta Prometheus, inside the gigawatt supercluster scaling to 1 million GPUs](https://www.storagereview.com/news/meta-prometheus-gigawatt-supercluster-1-million-gpus-gb200-gb300-nvl72)