Amazon Web Services (AWS), Amazon’s cloud computing arm, and Nvidia said on August 26 they will expand their AI infrastructure partnership.

The company will deploy 2 million additional Nvidia graphics processing units (GPUs) across AWS’s global infrastructure in 2027 and 2028.

This adds to AWS’s earlier plan to add more than 1 million GPUs starting in 2026, as demand for AI computing rises.

The expansion will also bring Nvidia’s Vera central processing unit (CPU)-based infrastructure to AWS and connect next-generation AWS Trainium chips with Nvidia’s NVLink Fusion technology.

This extends the partnership beyond GPU supply to chip-level interoperability.

NVLink is used to tightly connect accelerators with high-bandwidth, low-latency communication.

The companies also said they plan to build AI factories for the US government, including 100,000 GPUs on AWS’s secure infrastructure for federal and national-security workloads classified at Impact Level 6 and above.

The announcement adds to a broader buildout of AI infrastructure by hyperscale cloud providers.

🔗 Source: Amazon

Recent NVIDIA developments