Google Cloud is preparing to make the Nvidia Vera Rubin infrastructure generally available on its platform.

The latest generation Nvidia hardware will be generally available next month, and will later be offered to European customers also.

The hardware is offered via Google Cloud's A5X instances, which bring together the Nvidia Vera Rubin NVL72 rack systems, Nvidia ConnectX-9 SuperNICs, Google Cloud Virgo Network, and are cooled by Google's fourth-generation liquid cooling technology.

The Rubin platform comprises six chips in total – the NVLink 6 Switch, ConnectX-9 SuperNIC, BlueField-4 DPU, and Spectrum-6 Ethernet Switch, in addition to the Vera CPU and Rubin GPU. Vera is the successor to Nvidia’s Grace CPU, while Rubin will succeed Blackwell GPUs, with Nvidia claiming Rubin will be capable of achieving 5x and 3.5x inference and training performance, respectively, compared to Blackwell.

The NVL72 system - Nvidia's rack-scale offering that comprises 36 CPUs and 72 GPUs - differs from the Vera Rubin generation in that it is 100 percent liquid-cooled and features cable‑free modular tray designs, which Nvidia claims will allow installation times to be reduced from two hours to five minutes.

The search and cloud giant plans to scale up to as many as 960,000 Rubin GPUs across multiple sites.

The instances are also available as part of Google Cloud's AI Hypercomputer.

An early customer of the Vera Rubin hardware via Google Cloud is Ineffable Intelligence, a London-based tech company.

“We’re building superlearners that continually improve through their own experience,” said David Silver, founder and CEO of Ineffable Intelligence. “That demands both substantial compute and a tight connection between generating experience and training on it, so each improvement in the agent can quickly shape its next experience. Vera Rubin’s combination of compute performance and high-bandwidth connectivity, between the Vera CPU and Rubin GPUs and between the Rubin GPUs within the rack, allow us to push the scale and ambition of our reinforcement learning research further.”

In addition to the Vera Rubin availability, Google Cloud will be among the first cloud providers to offer Nvidia Confidential Computing at rack scale on bare metal instances such as A4X Max instances powered by Nvidia GB300 NVL72. Additionally, Google Cloud is making Confidential Computing generally available on G4 VMs powered by Nvidia RTX PRO 6000 Blackwell Server Edition GPUs.

Other cloud providers offering the Vera Rubin hardware include CoreWeave, which claimed to be the first to have a system up and running.