NVIDIA and AWS Collaborate to Bring AI to Production at Scale

NVIDIA Blog · Josiah Byers · 2026-06-24

NVIDIA and AWS announce an expanded collaboration introducing EC2 G7 instances with RTX PRO 4500 Blackwell GPUs, GPU-accelerated vector search as the default in Amazon OpenSearch Serverless via NVIDIA cuVS, and AWS achieving NVIDIA Exemplar Cloud status for GB300 training.

Open original ↗

Appears in

NVIDIA Expands Enterprise AI Ecosystem Across Cloud, Agents, and Industry Verticals

Extraction

Topics: gpu-infrastructurecloud-computingvector-searchai-inference

Claims

Amazon EC2 G7 instances powered by NVIDIA RTX PRO 4500 Blackwell GPUs deliver up to 4.6x AI inference performance and up to 2.1x graphics performance compared to G6 instances.
NVIDIA cuVS is now the default vector indexing compute choice in Amazon OpenSearch Serverless, enabling up to 10x faster vector indexing at a quarter of CPU-only cost.
AWS has achieved NVIDIA Exemplar Cloud status for GB300, certifying that AWS meets NVIDIA's reference architecture performance thresholds for large-scale training workloads.
The G7 instances support up to 8 GPUs, 256GB of total GPU memory, 700 Gbps EFA networking, and 7.6TB of local NVMe SSD storage.

Key quotes

vector indexing up to 10x faster at a quarter of the cost, compared with CPU-only builds — making billion-scale vector databases practical to build in under an hour.

Together, these advancements reinforce every layer of the AI infrastructure stack on AWS. The throughline is the same: production-grade AI infrastructure that performs at scale, without adding operational burden to the teams running it.