Amazon Web Services (AWS) has signed on to expand its fleet of Nvidia GPUs.

The company has agreed to deploy an additional 2 million GPUs across its global infrastructure, and the two companies will also deepen their work on AI factories, CPUs, networking, open models, data processing, and robotics.

– Nvidia

AWS will deploy the GPUs through 2027 and 2028. In addition, the cloud giant will bring Nvidia Vera CPU-based infrastructure to AWS, extend Nvidia NVLink Fusion with custom Nvidia high-bandwidth memory, and build AI factories for the US government with 100,000 GPUs on secure AWS infrastructure.

Nvidia's Nemotron models will also be offered via AWS, and the two will further robotics efforts with Amazon Robotics to adopt Nvidia's physical AI platform.