Meta Platforms has embarked on an ambitious data center build-out to power its AI ambitions. This includes Hyperion, a massive facility to be built in Louisiana that “will be able to scale up to 5 gigawatts over several years,” Meta CEO Mark Zuckerberg posted on Threads with a graphic illustrating how its planned footprint is nearly big enough to cover most of Manhattan. The first new plant to come online will be Prometheus in Ohio in 2026, Zuckerberg said, indicating his company will not be looking to OpenAI and Stargate partners to power Meta AI processing. Continue reading Meta Power Play Includes Data Centers of Up to 5 Gigawatts

Oracle will supply massive compute power to OpenAI as part of a new contract reported at $30 billion annually focused on accelerating Sam Altman’s ambitions for Stargate, the initiative to build U.S. data centers announced in January by President Trump as a matter of national security. OpenAI committed $500 billion over four years to the project. The Oracle deal involves an estimated 4.5 gigawatts of U.S. data center power consumption, equivalent to the power output of four-and-a-half nuclear reactors operating at full capacity — enough to power approximately 3.3 million U.S. households during that time. Continue reading OpenAI Deal with Oracle to Scale Up U.S. Cloud Infrastructure

Nvidia is rolling out DGX Cloud Lepton, a platform that connects AI developers with GPU access available through various cloud providers. Nvidia calls it “a compute marketplace” that offers tens of thousands of GPUs through a global network that features Nvidia Cloud Partners (NCPs). Among them: CoreWeave, Crusoe, Firmus, Foxconn, GMI Cloud, Lambda, Nebius, Nscale, Softbank Corp. and Yotta Data Services — offering Nvidia Blackwell and other architecture GPUs. Developers can tap into GPU compute capacity in specific regions for both on-demand and long-term computing, Nvidia says, adding that it expects leading cloud computing providers to eventually sign on. Continue reading DGX Cloud Lepton: Nvidia’s New GPU Compute Marketplace