Accelerating the agentic era: How Arm and Google Cloud co-design infrastructure for TPU 8i Max

What happened

Arm-based Google Axion host processors extend their role across Google’s TPU portfolio to deliver the high performance and energy efficiency required for demanding AI workloads. Google Cloud is pairing one of its newest TPU system s with Google Axion processors that are custom Arm Neoverse® -based CPUs (the general-purpose processor at the centre of a computer) . Google is a software and Internet company based in Mountain View, and its products and services include Software tools. Arm is based in Cambridge, and its products and services include software.

With TPU 8i Max, Google is extending its use of Axion CPUs as the host compute for its AI accelerators, building on Axion’s role in TPU 8t and TPU 8i. Google Axion provides the CPU foundation to support Google’s AI infrastructure across large-scale training, real-time reasoning, and increasingly demanding agentic (AI that carries out multi-step tasks rather than answering one question) workloads.

Google’s expansion to TPU 8i Max marks another significant milestone for Arm at the heart of the systems powering next-generation AI infrastructure. Advancing performance across the AI system AI performance depends on more than the accelerator.

Share Google Cloud is pairing​​ ​one of ​its newest TPU system​s​ ​with​ ​Google Axion processors that are custom Arm Neoverse® -based CPUs​.

CPUs prepare data, orchestrate workloads and execute the infrastructure software that keeps the entire system productive.

As agents reason, use tools and coordinate complex tasks, the demands on that CPU foundation continue to grow.