Daniel Romero(@HyperTechInvest):$AMD at the Citi Global TMT Conference “We haven't given a ton of details yet, but we do have some of our own internal silicon ambitions for ultra-low-latency inference that would fit into our architecture via chiplets.” The Taalas acquisition and the Cerebras partnership are just $AMD’s baby steps into ultra-low-latency inference The goal is to develop this silicon in-house going forward, and the trend we’ll likely see is increasingly specialized ASICs designed for specific workloads

2026.09.09 17:30

$AMD at the Citi Global TMT Conference “We haven't given a ton of details yet, but we do have some of our own internal silicon ambitions for ultra-low-latency inference that would fit into our architecture via chiplets.” The Taalas acquisition and the Cerebras partnership are just $AMD’s baby steps into ultra-low-latency inference The goal is to develop this silicon in-house going forward, and the trend we’ll likely see is increasingly specialized ASICs designed for specific workloads