Amazon Web Services (AWS) has launched a new AWS AI Factory offering that brings AI infrastructure to customers’ data centers.

Revealed at the company's Re:Invent 2025 conference in Las Vegas, the AI Factory offering sees AWS bring AI infrastructure, including Nvidia GPUs, Trainium chips, and AWS networking, storage, and databases to customers’ own data centers.

The dedicated infrastructure is operated exclusively for the customer, enabling governments and large organizations to scale AI projects while meeting compliance and sovereignty needs.

The factories operate like a private AWS Region, and give access to AWS managed services, including foundation models, while controlling where data is processed and stored.

Customers can opt for an Nvidia-AWS AI Factories integration, which gives them access to Nvidia hardware, full-stack Nvidia AI software, and the Nvidia computing platform. The AWS Nitro System, Elastic Fabric Adapter (EFA) petabit scale networking, and Amazon EC2 UltraClusters also support the Nvidia Grace Blackwell and the next-generation Nvidia Vera Rubin platforms. In addition, AWS is planning to make its future Trainium4 chips compatible with Nvidia NVLink Fusion.

“Large-scale AI requires a full-stack approach – from advanced GPUs and networking to software and services that optimize every layer of the data center. Together with AWS, we’re delivering all of this directly into customers’ environments,” said Ian Buck, vice president and general manager of Hyperscale and HPC at Nvidia.

“By combining Nvidia’s latest Grace Blackwell and Vera Rubin architectures with AWS’s secure, high-performance infrastructure and AI software stack, AWS AI Factories allow organizations to stand up powerful AI capabilities in a fraction of the time and focus entirely on innovation instead of integration.”

The AI Factory offering builds on AWS’ Project Rainier – in itself an AI Factor that was created for Anthropic based on Trainium2 chips – and is also the model being used for the company's partnership with Humain in Saudi Arabia.

Last month, AWS and Humain announced they had expanded their partnership to include the deployment of around 150,000 AI chips, including the Nvidia GB300s and Trainium chips.

“The AI factory AWS is building in our new AI Zone represents the beginning of a multi-gigawatt journey for Humain and AWS. From inception, this infrastructure has been engineered to serve both the accelerating local and global demand for AI compute,” said Tareq Amin, CEO of Humain.

“What truly sets this partnership apart is the scale of our ambition and the innovation in how we work together. We chose AWS because of their experience building infrastructure at scale, enterprise-grade reliability, breadth of AI capabilities, and depth of commitment to the region. Through a shared commitment to global market expansion, we are creating an ecosystem that will shape the future of how AI ideas can be built, deployed, and scaled for the whole world.”

AWS’ Re:Invent conference has also seen the launch of Trainium3 UltraServers, and the detailing of AWS’ planned Trainium4 chips.

AWS recently announced plans to spend $50bn expanding AI and HPC capacity for the US government.