AWS and NVIDIA Plan 2 Million Additional GPUs for AWS Infrastructure
Amazon Web Services and NVIDIA said on August 26 that AWS plans to deploy 2 million additional NVIDIA GPUs across its global infrastructure in 2027 and 2028. The expansion builds on AWS's earlier plan to add more than 1 million GPUs and adds Vera CPUs, networking, software, and federal AI-factory work to the companies' collaboration.
Amazon Web Services and NVIDIA said on August 26 that AWS plans to deploy 2 million additional NVIDIA GPUs across its global infrastructure in 2027 and 2028. The companies described the plan as an expansion of an earlier AWS commitment to add more than 1 million NVIDIA GPUs beginning in 2026.
What the announcement covers
AWS said the new capacity will include NVIDIA Blackwell Ultra, Rubin, and Rubin Ultra GPUs. The companies also said they are expanding work around NVIDIA Vera CPU-based infrastructure, networking, NVIDIA's Nemotron open models, GPU-accelerated data processing, and robotics.
The announcement includes plans to build AI factories for U.S. government workloads, including 100,000 GPUs on AWS secure infrastructure for federal and national-security use. Those plans are prospective, and the companies did not disclose commercial terms for the broader GPU deployment.
Why the timing matters
The 2027-2028 deployment window makes this a capacity plan rather than a statement that all of the hardware is available today. Reuters reported that NVIDIA's finance chief also described the additional two-million-GPU deployment on the company's earnings call.
For teams planning on AWS, the practical signal is that infrastructure availability will depend on more than accelerator counts. Delivery schedules, networking, power, and the availability of the named hardware generations will determine when capacity can serve workloads. The announcement also shows AWS and NVIDIA extending their work beyond GPU instances into data processing, model tooling, and physical-AI infrastructure.
Key Points
- 1AWS and NVIDIA plan to deploy 2 million additional NVIDIA GPUs across AWS global infrastructure in 2027 and 2028.
- 2The plan follows AWS's earlier commitment to add more than 1 million NVIDIA GPUs beginning in 2026.
- 3The expansion also covers Vera CPU infrastructure, networking, data processing, open models, robotics, and planned U.S. government AI factories.
Scoring Rationale
An official multiyear AWS-NVIDIA capacity plan with a precise deployment window, material GPU count, and implications for cloud AI infrastructure.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
