Capacity & Leasing

Data center capacity for lease in Texas and the Southeast

20–100 MW blocks on secured power. Standardized campuses, modular delivery, and NNN structures built for hyperscale and neocloud tenants.

The Offering

Standardized campuses, sized for the inference era

Phoenix delivers data center capacity in the range the market most needs and incumbents least serve: 20–100 MW blocks, leased on long-dated triple-net (NNN) terms to investment-grade hyperscale and credit-worthy neocloud tenants. Our repeatable 52 MW reference campus — built as modular 24 MW blocks — maps cleanly to modern GPU cluster deployments and compresses both cost and time-to-energization.

52 MW

Reference campus

A standardized critical-IT design, repeated across the platform for speed and cost discipline.

24 MW

Modular blocks

Phased delivery in increments sized to modern AI cluster architectures.

15-yr

NNN lease terms

Long-dated structures with contractual escalators, built for institutional counterparties.

Flagship

Texas capacity on the ERCOT grid

Our flagship North Texas campus — Project Panther — anchors the platform in the deepest pool of near-term grid power in the country. Served by Oncor at distribution level, the campus pairs secured interconnection with the fiber density the region is known for.

Project Panther — key specifications

  • Connected utility capacity 72 MW
  • Critical IT capacity ~52 MW
  • Configuration 3 × 24 MW modular blocks
  • Design density Supports ~75 kW/cabinet average
  • Cooling Air-cooled chillers + CRAC/CDU; high-density liquid ready
  • Fiber Seven carriers on route
  • Target energization 2028

Additional campuses across Texas, Florida, and the Carolinas follow the same reference design, giving tenants a repeatable expansion path across power-rich markets — without the multi-year interconnection queues of primary markets.

Use Cases

Built for cloud, compute, and AI inference workloads

Phoenix campuses are purpose-built for the fastest-growing segment of demand: latency-sensitive inference and cloud workloads that need to sit near users in metro-proximate, power-rich markets. Distributed-redundant topology, liquid-cooling readiness, and phased delivery let tenants match capacity to their deployment curve rather than over-committing on day one.

Need capacity on a real timeline?

Tell us your megawatts, density, and target dates — we'll show you where they fit on the platform.

Start the Conversation Explore the Platform