AI Data Center Power & Energy (2026)
The power ladder from one GPU to a whole data center: a single H100 draws 700W, a GB200 NVL72 rack ~120kW (over ten times a traditional rack), and large AI clusters reach hundreds of MW to GW scale. This table explains why AI is so power-hungry.
| Item | Power/value | Notes | Basis/source |
|---|---|---|---|
| NVIDIA H100 SXM5 | 700 W | Hopper flagship GPU | Official spec |
| NVIDIA H200 SXM | ~700 W | Hopper, 141GB HBM3e | Official spec |
| NVIDIA B200 (air, HGX/DGX) | 1,000 W | Blackwell, configurable TDP | Official spec (air config) |
| NVIDIA B200 (liquid, GB200) | 1,200 W | Full-spec liquid, ~500W over H100 | Media/GTC (configurable) |
| AMD Instinct MI300X | 750 W | 192GB HBM3, peak board | Official spec |
| AMD Instinct MI325X | 1,000 W | 256GB HBM3E, peak board | Official spec |
| Traditional rack (non-AI) | ~5–10 kW/rack | General enterprise/cloud rack | Industry typical (range) |
| GB200 NVL72 full rack | ~120 kW/rack | 72 Blackwell GPUs, liquid-cooled | NVIDIA design value (field 130–132kW) |
| PUE — industry average | 1.54 | Power usage effectiveness, ~1 is best | Uptime Institute 2024 survey |
| PUE — hyperscaler | 1.09–1.15 | Google fleet-wide TTM 1.09 | Operator self-reported |
| xAI Colossus (Memphis) | ~250 MW / 100k H100 | Single large training cluster | Media report (2024, fast-moving) |
Method & sources
Retrieved 2026-08-28. **Three bases, do not mix**: (1) **Chip TDP** from official specs, highest confidence — but NVIDIA Blackwell is "configurable TDP": the same B200 is 1,000W in air-cooled HGX/DGX and 1,200W in liquid-cooled GB200, labeled by configuration; (2) **Rack power** — two incomparable bases: traditional (non-AI) racks ~5–10 kW; GB200 NVL72 is NVIDIA's ~120 kW design value (field deployments reported 130–132 kW); "typical H100 rack 30–40 kW" is derived (DGX H100 ~10.2kW official × servers per rack), not a single official figure; (3) **PUE** — industry average 1.54 (Uptime Institute) vs hyperscaler 1.09–1.15 (Google fleet-wide 1.09) are different populations, not interchangeable or averageable. **Cluster-scale power is the weakest (low confidence)**: xAI Colossus 250MW is a 2024 report; GW-scale figures are SemiAnalysis-style planning estimates that can be stale within months ("as reported early 2026"). Official chip-spec PDFs are binary and not machine-readable verbatim; figures reflect consistent multi-source citation of official specs. The rack multiple (AI ~12–17x traditional) is derived and shifts with the chosen 7–10kW / 120–132kW endpoints. Sources: Uptime Institute, Google efficiency, NVIDIA/AMD specs, TweakTown, SemiAnalysis.
Source: https://datacenter.uptimeinstitute.com/rs/711-RIA-145/images/2024.GlobalDataCenterSurvey.Report.pdf
Retrieved: 2026-08-28
FAQ
- What does "AI Data Center Power & Energy (2026)" cover?
- It covers NVIDIA H100 SXM5, NVIDIA H200 SXM, NVIDIA B200 (air, HGX/DGX), NVIDIA B200 (liquid, GB200), AMD Instinct MI300X, AMD Instinct MI325X, Traditional rack (non-AI), GB200 NVL72 full rack, compared across: Item, Power/value, Notes, Basis/source.
- What are the sources and methodology?
- Retrieved 2026-08-28.
- When was this data last updated?
- The data was retrieved/updated on 2026-08-28.