IREN Scales AI Cloud to 150,000 GPUs
Amid ongoing cloud GPU shortages, IREN announced it is scaling its AI cloud capacity to 150,000 GPUs. The move aims to position the company as one of the world's largest suppliers for large-scale AI inference and training, highlighting the intense race to build out capacity.
The expansion involves the purchase of over 50,000 NVIDIA B300 GPUs, which will be deployed in phases throughout the second half of 2026. This move is expected to support an annualized run-rate revenue of over $3.7 billion by the end of 2026. IREN will house the new hardware in its existing air-cooled data centers in Mackenzie, British Columbia, and Childress, Texas. The company's strategy focuses on vertically integrated operations, controlling its own grid-connected land and power in regions with plentiful renewable energy. To finance this growth, IREN has secured $9.3 billion in funding over the past eight months through customer prepayments, convertible notes, and GPU leasing and financing. The company anticipates approximately $3.5 billion in additional capital expenditures for the second half of 2026 to cover the costs of GPUs, servers, storage, and networking equipment. This major scaling initiative positions IREN among the largest AI cloud infrastructure providers globally. The company already serves a growing client base that includes a $9.7 billion agreement with Microsoft, as well as multi-year contracts with other AI firms like Together AI, Fluidstack, and Fireworks AI.
