AMD and Cerebras Announce Disaggregated AI Inference Solution Combining AMD Helios and Cerebras Wafer-Scale Engine
AMD and Cerebras announced a technical partnership to deliver a disaggregated AI inference solution that combines AMD Helios rackscale systems with the Cerebras Wafer-Scale Engine. The joint solution is designed to deliver ultra-low-latency inference alongside higher throughput, with the two companies claiming up to 5x higher tokens per second per watt (T/s/W). Cerebras plans to deploy AMD Helios systems in its data centers, with the joint solution expected to be available initially through Cerebras Cloud in the...
Key takeaways
AMD and Cerebras announced a technical partnership for disaggregated AI inference, unveiled at the Advancing AI 2026 event.
The solution combines AMD Helios rackscale systems (high-throughput prompt engine) with the Cerebras Wafer-Scale Engine (ultra-low-latency token generation).
The companies claim up to 5x higher tokens per second per watt (T/s/W) versus a Cerebras WSE-only configuration, based on internal July 2026 modeling with the Kimi 2.6 1T Model.
Cerebras plans to deploy AMD Helios systems in its data centers.
Initial availability of the joint solution is expected through Cerebras Cloud in the second half of 2026.
No financial figures, revenue contribution, cost impact, or forward financial guidance were disclosed in the press release.
Quotes included from AMD Chair and CEO Dr. Lisa Su and Cerebras CEO and co-founder Andrew Feldman.
Key metrics
Primary source
TickerStar generated this brief from an official company, investor-relations, earnings-call, or regulatory source. Review the primary material before making an investment decision.
Open Investor relationsFollow AMD in real time
Get material earnings, filing, IR, and earnings-call alerts by email.