PartnershipJul 23, 2026, 01:47 PM
AMD & Cerebras Partner for Ultra-Low-Latency AI Inference Solution
AI Summary
AMD and Cerebras Systems have announced a technical partnership to introduce a new disaggregated AI inference solution. This collaboration combines AMD Helios rackscale solutions with the Cerebras Wafer-Scale Engine to achieve ultra-low latency and significantly increased throughput and efficiency for advanced AI applications. The joint solution is projected to deliver up to 5x higher tokens per second per watt and will be deployed by Cerebras in its data centers, with availability through Cerebras Cloud starting in the second half of 2026.
Key Highlights
- AMD and Cerebras Systems announced a technical partnership for a new AI inference solution.
- The solution combines AMD Helios rackscale solutions with the Cerebras Wafer-Scale Engine.
- Aims to deliver ultra-low latency and increased throughput for advanced AI applications.
- Expected to deliver up to 5x higher tokens per second per watt (T/s/W).
- Cerebras plans to deploy AMD Helios systems in its data centers.
- Joint solution will be available first through Cerebras Cloud in H2 2026.
Price Impact
More from AMD