---
type: "post"
id: "meta-detailed-prometheus-its-gigawatt-scale-supercluster-in-a2cf33"
slug: "meta-detailed-prometheus-its-gigawatt-scale-supercluster-in-a2cf33"
title: "Meta detailed Prometheus , its gigawatt-scale supercluster in New Albany, Ohio."
description: "Meta’s Prometheus account makes a useful point about frontier-model infrastructure: the unit of performance is the coordinated cluster, not an isolated GPU."
retrieval_nugget: "Meta’s Prometheus account makes a useful point about frontier-model infrastructure: the unit of performance is the coordinated cluster, not an isolated GPU."
published_at: "2026-10-19"
updated_at: "2026-10-19"
record_date: "2026-10-19"
date_kind: "published_at"
topics: ["compute","ai-infrastructure"]
source_urls: ["https://engineering.fb.com/2026/10/19/data-center-engineering/inside-prometheus-meta-gigawatt-scale-supercluster"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:meta-detailed-prometheus-its-gigawatt-scale-supercluster-in-a2cf33"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:5d212b3302b11b7a170588805002f6f5840edb162a4ac27fae8e1b9c42acf244","reviewed_at":"2026-10-19T20:07:41.592Z","source_evidence_count":1,"verified_claim_count":2,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":2,"implication_count":2,"watch_condition_count":2,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Large-scale model training depends on the reliability and coordination of the full compute system, not only accelerator peak throughput.","observed_facts":[{"text":"Meta says Prometheus’s first phase is complete and was used to train Muse models; the cluster currently comprises hundreds of thousands of GPUs across interconnected buildings, with a target of roughly one million.","source_urls":["https://engineering.fb.com/2026/10/19/data-center-engineering/inside-prometheus-meta-gigawatt-scale-supercluster"]},{"text":"Meta reports more than 85% goodput for Muse Wonder pre-training and more than 90% for reinforcement-learning runs, with interruption intervals later rising above 54,000 GPU hours after fleet changes and firmware fixes.","source_urls":["https://engineering.fb.com/2026/10/19/data-center-engineering/inside-prometheus-meta-gigawatt-scale-supercluster"]}],"mechanism":"A layered system combines interconnected buildings, a dedicated Ethernet back-end for GPU-to-GPU communication, successive NVIDIA GPU platforms, and cooling designed for high-density deployment; reliability improvements increase usable training time.","why_now":"Meta ties the completed first phase to training its Muse family and frames Prometheus as infrastructure for future frontier models, while the published figures make reliability measurable rather than merely asserting scale.","implications":["The first implication is that cluster-level reliability and network design can determine how much nominal compute becomes usable training time.","As capacity scales, operators will need to track operational performance across buildings and hardware generations, not just accelerator counts."],"evidence_boundary":"These are Meta-reported engineering metrics; the post does not independently validate them or provide a complete cost, energy, or comparative-performance analysis.","watch_conditions":["The first watch condition is whether Meta publishes comparable goodput and interruption metrics as Prometheus scales toward its stated million-GPU target.","Whether reported reliability gains persist across additional buildings and hardware generations."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied for this specific Meta Prometheus infrastructure report, so this begins a distinct analytical record."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/meta-detailed-prometheus-its-gigawatt-scale-supercluster-in-a2cf33/","markdown":"https://newruntime.com/posts/meta-detailed-prometheus-its-gigawatt-scale-supercluster-in-a2cf33.md","json":"https://newruntime.com/posts/meta-detailed-prometheus-its-gigawatt-scale-supercluster-in-a2cf33.json"}
---

# Meta detailed Prometheus , its gigawatt-scale supercluster in New Albany, Ohio.

## Retrieval answer

Meta’s Prometheus account makes a useful point about frontier-model infrastructure: the unit of performance is the coordinated cluster, not an isolated GPU.

Meta’s Prometheus account makes a useful point about frontier-model infrastructure: the unit of performance is the coordinated cluster, not an isolated GPU. Meta says the first phase in New Albany is complete and was used to train Muse models; the site now has hundreds of thousands of GPUs, with a stated target of roughly one million. The company reports more than 85% goodput for Muse Wonder pre-training and more than 90% for reinforcement-learning runs.

Thesis: Large-scale model training depends on the reliability and coordination of the full compute system, not only accelerator peak throughput. Mechanism: A layered system combines interconnected buildings, a dedicated Ethernet back-end for GPU-to-GPU communication, successive NVIDIA GPU platforms, and cooling designed for high-density deployment; reliability improvements increase usable training time. The first implication is that cluster-level reliability and network design can determine how much nominal compute becomes usable training time.

Why now: Meta ties the completed first phase to training its Muse family and frames Prometheus as infrastructure for future frontier models, while the published figures make reliability measurable rather than merely asserting scale. Evidence boundary: These are Meta-reported engineering metrics; the post does not independently validate them or provide a complete cost, energy, or comparative-performance analysis. The first watch condition is whether Meta publishes comparable goodput and interruption metrics as Prometheus scales toward its stated million-GPU target. The practical question is whether those gains persist as the cluster expands across hardware generations and buildings.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Compute](https://newruntime.com/topics/compute.md): Explore the compute topic hub.
- topic hub: [Ai Infrastructure](https://newruntime.com/topics/ai-infrastructure.md): Explore the ai-infrastructure topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)