Nvidia says 1GW AI factory costs $50 billion, unveils DSX POD
Nvidia CEO Jensen Huang told the GTC Berlin audience that a 1‑gigawatt AI data‑center facility would cost about $50 billion and could generate roughly $40 billion in annual revenue. He warned that a three‑month delay could forfeit about $10 billion of that revenue.
Key points
- 1GW AI factory price tag is about $50 billion and could earn $40 billion annually, Huang said
- DSX POD aims to cut deployment time from six months to as little as 14 days; units expected early 2027
- Nvidia shares rose up to 2 % in pre‑market trade after the announcement
The company introduced DSX POD, a new reference design that integrates compute, networking, storage, software, power and cooling into a single AI‑factory package. Nvidia says the system can shrink the time needed to move equipment through the supply chain and start token production from six months (or four‑to‑five months) to as little as 14 days. DSX POD units are slated for early 2027 availability.
Following the announcement, Nvidia shares rose up to 2 % in pre‑market trading and the ticker appeared among the top trending symbols on Stocktwits, with retail sentiment shifting from neutral to bullish.
The story so far
7 episodes →- Nvidia says 1GW AI factory costs $50 billion, unveils DSX PODthis story
Nvidia GTC Berlin Keynote: Jensen Huang Puts $50 Billion Price Tag On 1GW AI Factory Amid DSX POD Launch
finance.yahoo.com · 21 October 2026
Loading the full article…
This text was published by finance.yahoo.com and written by Prabhjote Gill. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
4sources- Nvidia announces plug and play Vera Rubin racks that can be shipped on a truckPress · CNBC Technology ·
- Vertiv and NVIDIA's AI Factory Design Unlocks 30% More Token OutputPress · 247wallst.com ·
- NVDA, ZM, SPCX Stocks Rise Premarket: Nvidia Announces AI Factory Pods, Zoom Gets Upgrade, SpaceX Target RisesPress · finance.yahoo.com ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Infineon and NVIDIA advance 800 VDC power systems for AI data centers · 1 src
- Intel and AMD collaborate on oneDNN optimizations for PyTorch · 2 src
- Google Cloud and NVIDIA make A5X instances generally available next month · 2 src
- Nokia and Nvidia target 50% spectrum efficiency by 2027 · 3 src
- Nvidia expands Confidential Computing with Confidential Bare Metal for model builders · 1 src
Comments
via GitHub Discussions