Advanced Micro Devices introduced the AMD Helion agentic AI rack on October 12, 2026, offering an open infrastructure blueprint designed for agentic workloads.
Rack Configurations and Specifications
AMD designed the Helion platform around an Open Compute Project spec-design. The blueprint supports flexible configurations, including liquid- and air-cooled versions.
It includes a CPU-only rack powered by 6th Gen AMD EPYC processors supporting up to 32,768 CPU cores for hyperscale deployments, as well as an all-in-one rack combining EPYC CPU nodes, AMD Instinct MI350P GPU nodes, and general-purpose compute for enterprises.
Availability and Ecosystem Support
AMD plans to contribute the Helion specification to the Open Compute Project community. Ecosystem partners including Cisco, Dell Technologies, Hon Hai Technology Group, HPE, Lenovo, QCT, Supermicro, and Wiwynn are collaborating on the platform.
AMD Helion configurations are currently under development and are planned for release in November 2026.
Key facts and where they come from
- AMD introduced the Helion agentic AI rack as an open infrastructure blueprint based on an Open Compute Project spec-design.
AMD (NASDAQ: AMD) today introduced the AMD Helion agentic AI rack, an open blueprint for agentic AI infrastructure based on an Open Compute Project spec-design
- A CPU-only rack configuration can support up to 32,768 CPU cores using 6th Gen AMD EPYC processors.
A CPU-only rack powered by 6th Gen AMD EPYC™ processors that can support up to 32,768 CPU cores.
- AMD Helion configurations are planned to be available in November 2026.
AMD Helion agentic rack configurations are currently under development with ecosystem partners and are planned to be available in November 2026.