Microsoft claims to be the first cloud to bring online an Nvidia Vera Rubin NVL72 system.

In a post on X, Microsoft CEO Satya Nadella shared an image of the latest generation Nvidia NVL72 system, writing: "We're the first cloud to bring up an Nvidia Vera Rubin NVL72 system for validation, another big step in building the next generation of AI infrastructure with Nvidia."

According to a Microsoft blog post, this was said to be done in a Microsoft "lab," while the Vera Rubin NVL72 will be rolled out to the company's Azure data centers "over the next few months."

With Nvidia's GTC event in full swing, several other clouds have also announced their intention to adopt the generation of AI hardware in the coming months.

The Nvidia Vera Rubin generation of chips has been in full production since the start of this year.

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, with Nvidia claiming Rubin will be capable of achieving 5x and 3.5x of inference and training performance, respectively, compared to Blackwell.

The NVL72 system - Nvidia's rack-scale offering that comprises 36 CPUs and 72 GPUs - differs for 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.

Google Cloud shared its own blog post, detailing plans for the AI platform.

According to the company, Google will be among the "first cloud providers" to offer the Vera Rubin NVL72 systems, aiming for them to be available in the second half of 2026. Google said that it will be integrated with its AI Hypercomputer offering, an AI-optimized infrastructure as a service platform.

Amazon Web Services (AWS), meanwhile, has said it will be deploying more than one million Nvidia GPUs, including Blackwell and Rubin GPU architectures, within the next 12 months.

Vultr, a privately-owned cloud provider, plans to offer an "optimized inference stack" based on the Rubin platform, that can be deployed on public, private, or sovereign clouds by customers.

On the neocloud front, Nebius revealed that as part of an up to $27bn AI capacity contract secured with Meta, the company would be launching one of the first "large-scale" deployments of the Vera Rubin chips.

Nscale is, under a contract with Microsoft, set to establish a 1.35GW cluster of Vera Rubin chips at its upcoming Monarch Compute Campus.

GMI Cloud has confirmed that it is planning to bring "signficant capacity" of the Vera Rubin NVL72 online across its global sovereign AI factories, with buildouts already underway.

CoreWeave, meanwhile, revealed plans to deploy the Rubin GPUs within its cloud platform back in January 2026. CoreWeave will also offer access to Vera as a standalone platform, currently planning to be the first cloud to do so. Lambda similarly revealed it would be deploying the Vera Rubin NVL72s in January, with availability planned in the second half of 2026.