SANTA CLARA, Calif
NVIDIA (NASDAQ: NVDA) and Amazon Web Services (AWS) today announced new advancements designed to help organizations deploy artificial intelligence applications at production scale with improved performance, efficiency, and operational simplicity.
The collaboration introduces Amazon EC2 G7 instances powered by NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, expands GPU-accelerated vector search capabilities in Amazon OpenSearch Serverless through NVIDIA cuVS, and highlights AWS’s achievement of NVIDIA Exemplar Cloud status for NVIDIA GB300 training workloads.
The new Amazon EC2 G7 instances are engineered to support AI inference, graphics, spatial computing, video processing, and data analytics workloads. Compared with previous-generation G6 instances, G7 offers significant gains in AI inference and graphics performance while enabling organizations to scale workloads efficiently through flexible GPU configurations.
AWS also announced that Amazon OpenSearch Serverless now uses GPU-accelerated vector indexing powered by NVIDIA cuVS as the default option for vector collections. This enhancement is designed to accelerate retrieval-augmented generation (RAG), semantic search, recommendation systems, and agentic AI applications while reducing infrastructure complexity and costs.
According to AWS and NVIDIA, the integration can deliver vector indexing speeds up to 10 times faster and at substantially lower cost compared with CPU-only approaches, helping organizations build and deploy large-scale AI retrieval systems more efficiently.
In addition, AWS has achieved NVIDIA Exemplar Cloud status for NVIDIA GB300 training workloads, demonstrating performance that meets NVIDIA’s reference architecture standards for large-scale AI training environments. The designation reflects ongoing engineering collaboration between the two companies to optimize cloud infrastructure for advanced AI applications.
“These advancements provide organizations with a stronger foundation for building, training, and deploying AI at scale,” the companies said. “By combining high-performance computing, accelerated data retrieval, and optimized training infrastructure, AWS and NVIDIA are helping customers move AI initiatives from development to production more efficiently.”
The latest innovations reinforce AWS and NVIDIA’s commitment to delivering production-ready AI infrastructure that supports enterprise-scale workloads while minimizing operational overhead.
For more information, visit the AWS and NVIDIA websites.