Armada has introduced Orion, a 10 MW member of its Galleon family of modular data centers aimed at distributed AI deployments where customers want to own and operate infrastructure on their own land and power. The company describes Orion as deployable “in months” and designed to scale from a single unit to multi-site builds reaching hundreds of megawatts.
Orion packages compute, networking, power, and cooling into a prefabricated six-module core. Armada lists support for up to 2,880 GPUs across 40 racks per unit, with the ability to achieve greater than N+1 redundancy. The unit is described as compatible with NVIDIA Blackwell and built with a “Vera Rubin-ready architecture.” Armada also states that units will be deliverable in months from deposit.
Ten megawatts per modular unit is a meaningful step up in density for operators trying to keep AI capacity close to specific power sources, jurisdictions, or latency domains without committing to a single centralized campus build. But 10 MW blocks also move the engineering conversation quickly into utility interconnects, on-site power distribution design, heat rejection strategy, and commissioning repeatability—areas where “modular” only pays off if the deployment process is truly standardized.
Orion is also tied to Armada Bridge, the company’s software platform for distributed infrastructure. Armada says Bridge provides multi-GPU orchestration across racks and sites, a model marketplace for deploying open and commercial models on sovereign infrastructure, and NVIDIA Token Factory enablement, with additional ISV and model partners providing services such as model compression, optimization, and model lifecycle management.
Separately, Fossefall has selected Armada as one of its technology partners for future AI factories in Norway, Sweden, and Finland, and plans to use five Leviathan Galleons. Fossefall’s stated goal is to build more than 1 GW of Nordic-owned AI capacity by 2030, and it says it will own and operate the GPU infrastructure inside the Leviathans. Armada lists each Leviathan at a maximum total IT load of 1.77 MW and says the design is intended to operate across a wide range of environmental conditions. Fossefall cites a cold inland climate as a factor supporting natural cooling and energy-efficient operations, and Armada states its closed-loop cooling system eliminates daily water consumption.
“Orion raises the ceiling on what a single deployment can deliver—ten megawatts per unit, infinitely scalable, in months instead of years,” said Dan Wright, co-founder and CEO of Armada.
“We selected Armada for its best-in-class technology, proven execution capabilities, and ability to significantly accelerate time to market,” said Øyvind Vesterdal, CEO of Fossefall.
Source: Armada
