AWS and Nvidia partner on Project Ceiba, a GPU-powered AI supercomputer

Amazon Web Services and Nvidia have announced an expansion of their alliance that includes plans to add supercomputing capabilities to AWS’s artificial intelligence (AI) infrastructure. The companies announced the news at the AWS re:Invent conference in Las Vegas.

The biggest of the initiatives is Project Ceiba, a supercomputer that will be hosted by AWS for Nvidia’s own research and development teams. It will feature 16,384 Nvidia GH200 Superchips and be capable of processing 65 exaflops of AI, the companies said. The Project Ceiba supercomputer will be integrated with a number of AWS services, including Amazon Virtual Private Cloud (VPC) encrypted networking and Amazon Elastic Block Store high-performance block storage.

Nvidia plans to use the supercomputer for research and development to advance AI for LLMs, graphics and simulation, digital biology, robotics, self-driving cars, Earth-2 climate prediction and more.

New Amazon EC2 G6e instances featuring Nvidia L40S GPUs and G6 instances powered by L4 GPUs are also in the works, AWS announced. L4 GPUs are scaled back from the Hopper H100 but offer much more power efficiency. These new instances are aimed at startups, enterprises, and researchers looking to experiment with AI.

Nvidia also shared plans to integrate its NeMo Retriever microservice into AWS to help users with their development of generative AI tools like chatbots. NeMo Retriever is a generative AI microservice that enables enterprises to connect custom LLMs to enterprise data, so the company can generate proper AI responses based on their own data.

“Generative AI is transforming cloud workloads and putting accelerated computing at the foundation of diverse content generation,” said Jensen Huang, founder and CEO of Nvidia, in a statement. “Driven by a common mission to deliver cost-effective, state-of-the-art generative AI to every customer, Nvidia and AWS are collaborating across the entire computing stack, spanning AI infrastructure, acceleration libraries, foundation models, and generative AI services.”