Aws and Nvidia expand AI infrastructure with additional 2 million gpus

Amazon Web Services (AWS) has announced plans to deploy an additional 2 million NVIDIA GPUs across its global infrastructure between 2027 and 2028. This follows a previous commitment made in March 2026 to deploy 1 million NVIDIA GPUs, bringing the total number of GPUs to 3 million. The expansion includes the integration of NVIDIA Groq LPU into AWS's cloud services. AWS also plans to offer infrastructure based on NVIDIA Vera CPU, integrating its own chips with NVIDIA NVLink Fusion and NVHBM. The partnership between AWS and NVIDIA spans multiple areas including artificial intelligence applications in manufacturing, network connectivity, models, and physics.