AMD showcases “Helios” rack-scale platform built on the Open Compute Project Open Rack for AI, introduced by Meta.

At the Open Compute Project (OCP) Global Summit 2025, AMD showcased its “Helios” rack-scale platform, a reference design compliant with the new Open Rack Wide (ORW) standard contributed by Meta. The Helios platform integrates advanced AMD technologies, including Instinct GPUs and EPYC CPUs, enabling the development of infrastructure tailored for next-generation Artificial Intelligence (AI) workloads.

By adopting the ORW specification, Helios offers a scalable solution focused on meeting the increasing demands of AI across gigawatt-scale data centers. This architecture emphasizes open standards, promoting interoperability and flexibility within AI deployments while maintaining high performance.

Forrest Norrod, executive vice president and general manager of AMD’s Data Center Solutions Group, emphasized the importance of open collaboration in scaling AI infrastructure efficiently. He noted that Helios combines AMD’s leading-edge technologies into a platform designed to support the unique requirements of future AI applications.

The Helios platform is built around multiple open compute standards, allowing quick customization and deployment by original equipment manufacturers, original design manufacturers, and hyperscalers. This collaboration with the OCP community aims to facilitate efficient infrastructure for AI and High performance computing (HPC) workloads globally.