IREN announced new multi-year AI cloud agreements with leading AI developers worth approximately $2.8 billion in total contract value (TCV), prompting the company to increase its year-end 2026 AI Cloud annualized run-rate revenue (ARR) target from $3.7 billion to more than $4.0 billion. The company said approximately 85% of the new ARR target is now under contract, reflecting continued demand for GPU infrastructure supporting AI training and inference.
The expanded customer roster includes Microsoft, NVIDIA, Perplexity, Figure AI, Together AI, Fluidstack, Fireworks AI, Fal AI, Hume AI, along with an additional unnamed AI developer. IREN said it continues to allocate capacity selectively as demand from hyperscalers, frontier AI labs, enterprises, and AI startups exceeds its available and planned infrastructure. The company expects to deliver 480 MW of AI cloud capacity during 2026 and targets 1.2 GW by 2027 through its global expansion program.
IREN also highlighted improving contract economics. Recent agreements include customer prepayments covering approximately 45% of the associated GPU capital expenditures, lowering the company’s upfront funding requirements. Across its customer portfolio, contracts now carry a weighted average term of roughly four years. As of June 30, 2026, IREN reported approximately $7.6 billion in cash and cash equivalents, including $1.7 billion of restricted cash related to GPU financing for its previously announced Microsoft deployment.
• AI Cloud ARR target increased from $3.7 billion to more than $4.0 billion.
• Approximately 85% of targeted ARR is now under contract.
• New contracts represent $2.8 billion in total contract value.
• Customer list includes Microsoft, NVIDIA, Perplexity, Figure AI, Together AI, Fluidstack, Fireworks AI, Fal AI, Hume AI, and another AI developer.
• Company expects 480 MW of AI cloud capacity by the end of 2026.
• Long-term target reaches 1.2 GW of AI cloud capacity during 2027.
• Recent contracts include customer prepayments covering about 45% of GPU capital costs.
• Weighted average customer contract term is approximately 4 years.
• Cash and cash equivalents totaled approximately $7.6 billion at June 30, 2026.
Daniel Roberts, Co-Founder and Co-CEO of IREN, said:
“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.”
🌐 Analysis
IREN continues its transformation from a bitcoin mining operator into a large-scale AI infrastructure provider. The announcement follows several major financing transactions and the company’s Microsoft GPU deployment, reinforcing a strategy centered on long-term AI cloud contracts backed by dedicated GPU clusters and renewable-powered campuses. The inclusion of customer prepayments is notable because it reduces capital intensity while improving funding visibility for future deployments.
The announcement also reflects the increasingly competitive market for AI cloud infrastructure. Companies such as CoreWeave, Crusoe, Lambda, Nebius, TensorWave, and traditional hyperscalers continue expanding GPU capacity to satisfy demand for large-scale AI training and inference. IREN’s growing roster of enterprise and frontier AI customers suggests that independent GPU cloud providers remain an important complement to hyperscale cloud platforms as global AI compute demand continues to outpace available infrastructure.
