AIThe Informationabout 1 hour ago
AMD and Cerebras Partner on AI Inference Platform
1 min read
AMD and Cerebras Systems announced a platform combining AMD's EPYC CPUs and Instinct MI400-series accelerators in Helios racks with Cerebras' Wafer-Scale Engine. The disaggregated design assigns prompt processing to AMD hardware and token generation to Cerebras, aiming for up to 5X higher tokens per second per watt. Availability through Cerebras Cloud is expected in the second half of 2026.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAMD
