Nvidia signs $50 billion lease for Hut 8 Texas data center

Nvidia has committed up to $50 billion to lease Hut 8's 1-gigawatt Texas data center, the largest known single-tenant AI infrastructure deal by value.

Nvidia signed leases worth as much as $50 billion for Hut 8's Beacon Point campus in Texas, the Financial Times reported Tuesday, citing five people familiar with the deal. The 1-gigawatt facility represents one of the largest single-tenant data center commitments in history, underscoring the scale of capital Nvidia is deploying to secure computing capacity for its AI customers.

"The scale of this commitment reflects the structural demand for AI compute that we see across our customer base," said a person familiar with Nvidia's strategy, speaking on condition of anonymity because the terms are private. Hut 8 disclosed last week that the base-term contract value was $19.6 billion over 15 years, rising to $50.2 billion if all renewal options are exercised, but did not name the tenant at the time, describing it only as an "existing investment-grade customer."

The 1-gigawatt capacity would be enough to power roughly 850,000 average U.S. homes, based on Energy Information Administration data showing monthly household consumption of 865 kilowatt-hours. Nvidia could sublease portions of the space to its "neocloud" partners — companies that buy its graphics processing units and resell AI cloud computing services, according to the FT report. The deal comes as Nvidia also discusses providing $250 billion to help fund a separate 10-gigawatt OpenAI data center in Ohio, the Wall Street Journal has reported, a project that could exceed $500 billion in total cost including chips.

The Texas lease signals that Nvidia is moving beyond selling chips to directly controlling the physical infrastructure where those chips operate, a strategy that tightens its grip on the AI supply chain. Hut 8, a bitcoin miner that has pivoted to AI data center development, gains a marquee tenant that validates its Beacon Point project. For hyperscaler competitors such as Microsoft, Amazon and Google — each racing to build their own AI capacity — the deal shows Nvidia is willing to spend tens of billions to ensure its GPU ecosystem has the power and cooling capacity needed for next-generation training clusters. Nvidia shares have more than doubled over the past year as the company's data center revenue surged past $100 billion annually, though the stock trades at roughly 35 times forward earnings, a premium that reflects expectations of sustained AI infrastructure spending.

This article is for informational purposes only and does not constitute investment advice.