NVIDIA Vera Rubin infrastructure on Google Cloud heads to general availability

What is A5X?

Google Cloud and NVIDIA said A5X instances built on Vera Rubin infrastructure will reach general availability next month. The system combines NVIDIA Vera Rubin NVL72, ConnectX-9 SuperNICs, Google Cloud's Virgo network and the company's fourth-generation liquid-cooling technology.

The goal is to bring demanding agentic inference and large-scale reinforcement learning workloads onto one platform. Google Cloud plans to scale its Vera Rubin-based infrastructure to as many as 960,000 Rubin GPUs across multiple sites.

First customer and use cases

Ineffable Intelligence, the London-based company founded by AlphaGo creator David Silver, was the first customer to access A5X. It is using the platform to advance "superlearner" systems that learn continuously from experience. Silver said the high-bandwidth connectivity between the Vera CPU and Rubin GPUs, and among GPUs within the rack, lets the company push the scale of its reinforcement learning research further.

Marketing company WPP also demonstrated its creative agent reference architecture on A5X using Nemotron models and OpenShell. In initial tests, workloads that typically required a multi-node cluster ran on a single A5X instance.

Privacy and efficiency

Google Cloud will be among the first providers to offer rack-scale NVIDIA Confidential Computing on bare-metal instances such as A4X Max powered by GB300 NVL72. Confidential Computing is also reaching general availability on G4 VMs powered by RTX PRO 6000 Blackwell GPUs for confidential inference and fine-tuning.

On efficiency, the two companies are exploring NVIDIA DSX MaxLPS capabilities for Vera Rubin-based infrastructure.