By Luke James
Publication Date: 2025-12-02 18:20:00
AMD and HPE are Expanding your long-term partnership with a new deal that will add the AMD Helios rack-scale AI architecture to HPE’s product portfolio starting in 2026. This will give Helios its first major OEM backing and position HPE to ship full 72-GPU AI racks built around next-generation Instinct MI455X accelerators, new EPYC “Venice” CPUs, and a scalable Ethernet-based fabric developed with Broadcom.
Helios has been AMD’s reference design for an open rack-level AI platform since its introduction earlier this year. It uses Meta’s Open Rack Wide mechanical standard and combines MI450 series GPUs, Venice CPUs, and Pensando networking hardware within a liquid-cooled double-wide chassis.
AMD has targeted up to 2.9 exaFLOPS of FP4 compute per rack with the MI455X generationalong with 31TB of HBM4 and an expandable topology that exposes each GPU as part of a single module. HPE will implement that design with a purpose-built HPE Juniper switch that supports…
