AI Infrastructure · Santa Clara, California
AMD Targets November Availability for Helion Agentic AI Rack Designs
AMD’s October 12 Helion announcement outlines CPU-only and combined CPU/GPU rack designs, with partner configurations planned for November.
AMD introduced Helion on October 12 as an open rack blueprint for agentic AI, with partner configurations planned to become available in November 2026.
The design accommodates CPU-only racks for hyperscale agent execution and integrated CPU/GPU systems for enterprise deployments. AMD describes both air-cooled and liquid-cooled options, allowing configurations to reflect workload and facility requirements.
One proposed CPU rack supports up to 32,768 cores using sixth-generation EPYC processors. Another combines EPYC nodes with Instinct MI350P GPU nodes. These are vendor-described configurations, not independently validated operating deployments.
AMD plans to contribute the specification to the Open Compute Project. Buyers should confirm final partner availability, power envelopes and cooling requirements before using the announcement as a procurement specification.
Original publisher (October 12, 2026). Source checked October 19, 2026. · AI Infrastructure.
Key details
- CPU-only capacity
- Up to 32,768 CPU cores using 6th Gen AMD EPYC processors
Why it matters
CPU-only and CPU/GPU options let operators match agent execution and inference to power, cooling and cost constraints. Planned availability is not evidence of a customer order, facility investment or open procurement.
Status and next steps
AMD said Helion rack configurations were under development with ecosystem partners and planned for availability in November 2026.
Review the official next step →Project contacts and next steps
A direct project contact has not been verified for this article. Consult the linked source organization for the appropriate office.
Primary sources
- AMD: Introducing Helion agentic AI rack · October 12, 2026