NVIDIA Corporation

NVIDIA entered into multiple residual value guaranties (collectively, the "Agreements") with SB Energy (the "Lessor") relating to leases for approximately 4.25 gigawatts of IT load in the aggregate at the Portsmouth Site. ... NVIDIA's aggregate payment obligation is cumulatively capped at $105 billion for its initial commitment under the Agreements.

Extension: NVIDIA can provide credit support to secure approximately an additional 3.8 gigawatts at the Portsmouth Site, exercisable in its sole discretion.

The Company and NVIDIA entered into a collaboration framework to expand their long-standing complementary relationship to advance AI adoption at global scale and to accelerate the Company's buildout of more than 5 gigawatts of artificial intelligence factories by 2030 developed and operated by the Company using NVIDIA's computing platform technology.

Blackwell GPUs procured – 1.3k B200s & 1.1k B300s to be installed at Prince George over the coming months, lifting total fleet to 4.3k NVIDIA GPUs and unlocking next-gen training and inference at scale

IREN today announced the purchase of 2.4k next-generation NVIDIA Blackwell B200 and B300 GPUs for ~$130m, including fit-out costs. ... The order (1.3k B200 and 1.1k B300 GPUs) will be installed over the coming months at IREN's Prince George, BC campus

HIVE Expands NVIDIA Chip Suite to Power AI Boom with $30 Million H100 and H200 GPU Clusters to be Deployed in Quebec ... announces the acquisition and deployment of an additional cluster of cutting-edge NVIDIA H100 GPUs ... NVIDIA H100 cluster: Comprised of 248 GPUs in 32 nodes connected with Infiniband, this cluster has arrived on site in Quebec, has been fully configured and will be operational before the end of 2024. ... NVIDIA H200 cluster: Comprised of 508 GPUs in 64 nodes connected with Infiniband, this cluster has been shipped and is expected to arrive in Quebec in early January 2025, with installation and customer deployment planned for calendar Q1

On September 16, 2024, Iris Energy Limited (dba IREN) (the “Company”) announced its purchase of an additional 1,080 NVIDIA H200 GPUs to expand its AI Cloud Services business. ... The Company is purchasing 1,080 of the latest-generation NVIDIA H200 GPUs for $43.9m, with delivery and commissioning expected during 4Q 2024.

During the second quarter, the Company purchased a 128-GPU cluster from NVIDIA, financed by a leading OEM.

Tripling AI cloud services business ... Fast-tracking AI cloud services business to 816 NVIDIA H100 GPUs with the purchase of 568 additional GPUs ... Delivery and commissioning of additional GPUs expected during Q2 2024

Backed by Nvidia who owns a ~11.5% equity stake and is obligated to purchase up to $6.3Bn of residual cloud capacity through April 2032

On September 9, 2025, CoreWeave, Inc. (the “Company”) and NVIDIA Corporation (“NVIDIA”) entered into a new order form (the “Order Form”) under the existing Master Services Agreement (“MSA”) dated as of April 10, 2023, which has an initial value of $6.3 billion, that establishes an arrangement with respect to the sale by the Company of reserved cloud computing capacity to its customers and provides NVIDIA access to any residual unsold cloud computing capacity. Under the terms of agreement, in instances where the Company’s datacenter capacity is not fully utilized by its own customers, NVIDIA is obligated to purchase the residual unsold capacity through April 13, 2032, subject to any termination described below and satisfaction of delivery and availability of service requirements.