IREN Limited has announced $2.8 billion in new multi-year contracts with leading AI developers, raising its year-end revenue target from $3.7 billion to over $4 billion, with approximately 85% now under contract. The announcement reflects surging demand for large-scale computing infrastructure as artificial intelligence companies race to build the data centers needed for training and running advanced AI models.

Who Is Betting on IREN's AI Infrastructure?

IREN's customer base now spans some of the most prominent names in AI and cloud computing. The company has secured contracts with Microsoft, NVIDIA, Perplexity, Figure AI, Together AI, Fluidstack, Fireworks AI, Fal AI, Hume AI, and a new unnamed leading AI developer. These customers are using IREN's platform for both bare metal services, which give direct access to raw computing hardware, and managed cloud services, which handle infrastructure management on the customer's behalf.

The diversity of IREN's customer base is notable. Rather than relying on a single hyperscaler, the company has deliberately spread its capacity across multiple AI developers, enterprises, and frontier research labs. This strategy reduces risk while positioning IREN as a critical infrastructure provider across the AI ecosystem.

How Is IREN Scaling Its Data Center Capacity?

IREN's growth trajectory is remarkable. The company expanded from approximately 3 megawatts of self-built AI cloud capacity just 12 months ago to 480 megawatts planned for delivery by the end of 2026, with 1.2 gigawatts targeted for 2027. To put this in perspective, a megawatt of computing power can support thousands of simultaneous AI model training runs or inference requests, depending on the workload.

• Current Expansion: IREN is delivering 480 megawatts of AI cloud capacity in 2026, a 160-fold increase from its capacity just one year prior.
• Future Plans: The company targets 1.2 gigawatts of capacity by 2027, representing continued aggressive expansion to meet customer demand.
• Geographic Diversification: IREN's infrastructure spans North America, Europe, and Asia-Pacific, leveraging renewable-rich regions to support its power-intensive operations.

What makes IREN's approach distinctive is its vertical integration. Rather than simply renting server space, IREN owns and operates its own data centers and controls access to grid-connected land and power resources. This gives the company more control over costs, reliability, and the ability to customize infrastructure for AI workloads.

What Are the Financial Terms of These Contracts?

The financial details reveal how seriously AI companies are committing to infrastructure expansion. Recent contracts include customer prepayments representing approximately 45% of the associated graphics processing unit (GPU) capital expenditure, significantly reducing IREN's net funding requirement for those deployments. In other words, customers are paying upfront for a substantial portion of the hardware costs before the systems are even built.

The weighted average contract term across IREN's portfolio is approximately four years, indicating long-term commitments from customers. This stability allows IREN to plan capacity expansions with confidence and secure financing for infrastructure buildout. As of June 30, 2026, IREN held approximately $7.6 billion in cash and cash equivalents, including $1.7 billion in restricted cash tied to GPU financing for Microsoft contracts.

"Our vertically integrated AI Cloud platform is scaling at pace. In the past 12 months we have expanded from approximately 3MW of self-built AI Cloud capacity to 480MW being delivered this year, with 1.2GW targeted for 2027, broadening our customer base across hyperscalers, enterprises and AI developers," said Daniel Roberts, Co-Founder and Co-CEO of IREN.

Daniel Roberts, Co-Founder and Co-CEO of IREN

Why Does This Matter for the AI Industry?

The scale of IREN's expansion underscores a critical bottleneck in AI development: power and computing infrastructure. As AI models grow larger and more capable, they require exponentially more computational resources. Training a state-of-the-art large language model can consume millions of dollars worth of computing power, and running inference at scale for millions of users demands constant, reliable access to GPU clusters.

IREN's success in securing $2.8 billion in new contracts signals that demand from AI developers continues to exceed available capacity. The company explicitly noted that "demand from hyperscalers, enterprises, AI developers and frontier labs continues to exceed IREN's available and planned capacity". This supply constraint is driving up prices for computing resources and creating opportunities for infrastructure providers willing to invest in large-scale data centers.

The prepayment terms in recent contracts also reveal customer confidence. When AI companies are willing to pay 45% of hardware costs upfront, it demonstrates their conviction that they will need this computing capacity and that the pricing is acceptable. This contrasts with more cautious purchasing patterns and suggests the AI infrastructure market is moving from experimental to mission-critical.

What Does IREN's Strategy Tell Us About AI's Future?

IREN's focus on renewable-rich regions and vertically integrated operations reflects broader industry trends. As AI companies face pressure to reduce their carbon footprint, infrastructure providers that can offer computing power backed by renewable energy have a competitive advantage. IREN's portfolio of grid-connected land and power in renewable-rich regions positions it to meet both performance and sustainability demands.

The company's selective approach to customer allocation is also telling. Rather than signing contracts with every interested party, IREN is prioritizing diversification across its customer base and platform layers. This strategy protects the company from over-dependence on any single customer while ensuring it can support a broad ecosystem of AI applications, from generative media and robotics to AI search and model development.

Looking ahead, IREN's 2026 annualized run-rate revenue target of over $4 billion represents a significant milestone for the company. However, the company cautioned that this target reflects internal assumptions regarding GPU models, contracting, utilization, and pricing, and that revenue is subject to successful commissioning, testing, and customer acceptance of GPUs in the months following each data center's delivery. In other words, while the contracts are signed, the actual revenue depends on IREN's ability to execute on its aggressive expansion timeline.