Amazon Web Services is increasing its strategic allocation of Nvidia hardware to support next-generation infrastructure.

Nvidia (NVDA) and Amazon (AMZN) are dramatically expanding their artificial intelligence infrastructure partnership, with Amazon Web Services (AWS) planning to deploy two million additional Nvidia GPUs across its global network in 2027 and 2028. This unprecedented scale grants Nvidia unusually long demand visibility across multiple product generations while reinforcing Amazon’s determination to keep AWS competitive as AI workloads increasingly drive cloud spending.

The expanded agreement covers Nvidia’s Blackwell Ultra, Rubin, and Rubin Ultra GPUs, alongside deeper integration across CPUs, networking, open models, data processing, and robotics. AWS and Nvidia also plan to build AI factories for the U.S. government, dedicating 100,000 GPUs specifically to federal and national-security workloads.

The partnership extends well beyond raw compute. Amazon Robotics will adopt Nvidia’s physical AI platform for warehouse automation, while AWS plans to launch infrastructure built around Nvidia Vera CPUs. The companies will also deepen support for Nvidia’s Nemotron models through Amazon Bedrock and SageMaker.

The announcement follows another blockbuster quarter for Nvidia. Fiscal second-quarter revenue surged 106% year over year to $96.2 billion, with Data Center revenue jumping 117% to $89 billion. Nvidia guided for approximately $108 billion in third-quarter revenue, explicitly excluding any China Data Center compute sales from its forecast.

For Nvidia investors, Amazon’s two-million-GPU commitment is significant because it stretches demand visibility beyond the current Blackwell cycle and into the Rubin architecture, reducing concerns that hyperscaler AI spending could peak prematurely. The central question is whether Amazon’s expanded deployment will trigger a broader wave of multi-year commitments across the hyperscale sector.

Investors should closely monitor Nvidia’s Data Center growth, the Rubin production ramp, gross margin trends, and the execution pace at which AWS brings the additional capacity online. With Nvidia expecting a third-quarter gross margin of roughly 74%, maintaining profitability while scaling successive architectures remains critical.

Strategically, the deal raises the stakes around AWS monetization and capital efficiency for Amazon. For Nvidia, it strengthens the case that AI infrastructure spending is evolving into a sustained, multi-year buildout rather than a single-cycle hardware surge.