aiqbench takeChips & Infra

GlobalFoundries' MIPS Acies platform targets edge AI inference by optimizing tokens-per-watt rather than raw compute power, addressing the critical energy constraints of deploying intelligence in industrial and embedded systems. By leveraging open RISC-V standards and a software-first architecture, this approach aims to simplify developer adoption while maximizing efficiency for real-world physical applications.

Source coveringQuantum Zeitgeist

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

GlobalFoundries’ MIPS Acies platform speeds up AI inference at the edge

Quantum ZeitgeistSeptember 8, 20263 min read

MIPS, by GF (MIPS), launched three workload-native platforms—MIPS Acies, MIPS Actus, and MIPS Aegis—to bring Physical AI capabilities to industrial machines, transportation platforms, and embedded systems. These platforms will enable commercial and ecosystem development for lowering modern AI onto tailored platforms with class-leading energy efficiency.

GlobalFoundries’ MIPS launched three new platforms, Acies, Actus, and Aegis, on September 2, 2026, each designed for distinct Physical AI applications ranging from inference to mission-critical systems. The company intends these workload-native platforms to accelerate the development of energy-efficient AI for industrial machines and embedded systems. “As intelligence moves into machines, vehicles, and other physical systems, compute platforms need to…

Excerpt shown under the publisher's RSS terms · 60 words · full text and figures are on the publisher's site

Read at Quantum Zeitgeist ↗Ask about this article