Nvidia signs $50 billion lease for Texas data center


Nvidia has signed leases worth up to $50 billion for Hut 8’s Beacon Point data center campus in Texas, according to the Financial Times and Reuters, marking one of the largest infrastructure commitments in the AI buildout race. The base-term contract value is $19.6 billion over 15 years, with renewal options that could push the total value to $50.2 billion over 30 years, Hut 8 announced on July 20.

The facility, located in Nueces County, is designed around Nvidia’s DSX reference architecture, a full-stack AI factory framework that combines Nvidia’s technology with equipment from partner companies to enable gigawatt-scale AI infrastructure. Beacon Point will deliver 704 megawatts of IT capacity at full build-out, supported by 1,000 megawatts of utility capacity secured through an interconnection agreement with AEP Texas.

A sprawling industrial data center campus under construction, massive steel structures and cooling systems visible across a flat Texas landscape, cranes and equipment scattered across the site, late afternoon light casting long shadows, scale and ambition evident | data center construction Texas

Nvidia could sublease the property to its “neocloud” partners—companies that buy Nvidia GPUs and resell AI cloud computing services—according to the Financial Times. An Nvidia spokesperson confirmed the company is “working with ecosystem partners to accelerate the deployment of efficient AI infrastructure through the DSX AI factory architecture,” without directly confirming the lease deal.

The transaction reflects Nvidia’s broader strategy to anchor the physical infrastructure powering AI. In recent months, Nvidia has invested billions in companies developing data center optics and power technology, committed roughly $100 billion in financing backstops for OpenAI’s data center buildout, and broken ground on a $700 million “AI factory” in Fort Worth. The company has also signed similar large-scale leases at other sites, signaling a shift from pure chip sales toward shaping the entire ecosystem of AI deployment.

A close-up view of high-density server racks in an AI data center, glowing LED indicators and cooling tubes visible, equipment stacked vertically to maximize power density, blue and red light from monitoring systems reflecting off metal surfaces, precision engineering | AI server rack cooling system

Hut 8 financed the Beacon Point project through junk-bond offerings and investment-grade debt, demonstrating how the AI infrastructure boom is attracting capital across credit markets. The company structured the lease as a triple-net (NNN) agreement, meaning the tenant covers operating expenses, taxes, and insurance. Hut 8 expects the facility to contribute $655 million per year in net operating income once stabilized, with initial Phase 2 delivery expected in Q2 2028.

The Beacon Point deal underscores the scale of AI infrastructure demand. Major hyperscalers—including OpenAI, Google, Microsoft, and Meta—are collectively planning to spend nearly $700 billion on data center projects in 2026 alone, according to TechCrunch. Gigawatt-scale facilities have become the new standard for training large language models, and securing power and land has become as critical as securing chips. Nvidia’s willingness to backstop large leases and shape facility design through its DSX architecture signals the company’s confidence in sustained AI compute demand and its central role in the infrastructure supply chain.

Sources

  • Reuters — Nvidia lease deal, Beacon Point specifications, and contract value details
  • Financial Times — Nvidia as tenant, neocloud sublease potential
  • PRNewswire (Hut 8 press release) — Beacon Point facility design, lease terms, capacity, NOI contribution, and timeline
  • TechCrunch — Hyperscaler data center spending forecast for 2026

Give your feedback

Be the first to rate this post
or leave a detailed review



ECIKS.org is an independent media. Support us by adding us to your Google News favorites:

Post a comment

Publish a comment