22,000 GPUs: Inflection AI Building 22 exaFLOPS Generative AI Cluster

Palo Alto-based startup Inflection AI yesterday said it is building the world’s largest AI cluster comprised of 22,000 NVIDIA H100 Tensor Core GPUs that will deliver 22 exaFLOPS performance. The company also said it has raised $1.3 billion in a funding round led by Microsoft, Reid Hoffman, Bill Gates, Eric Schmidt and new investor NVIDIA, bringing total funding raised by the company to $1.525 billion.

Along with partners CoreWeave, which is a GPU cloud provider (see recent Coreweave coverage), and NVIDIA, Inflection AI said it will develop the  cluster for training and deployment of large-scale generative AI models. “Combined, the cluster develops a staggering 22 exaFLOPS in the 16-bit precision mode, and even more if lower precision is utilized,” the company said in its announcement, adding that if the cluster were entered in the recent TOP500 list of supercomputers, it would be 2nd on the list “despite being optimized for AI – rather than scientific – applications.”

The rollout of the cluster is under way, the company said, and cited its performance in the recent MLPerf benchmark.

The cluster will support Inflection AI’s Pi, which stands for “personal intelligence,” based on a large language model designed for people to interact with AI “in the most simple, natural way and receive fast, relevant and helpful information and advice,” the company said.

Pi “is designed to be a kind and supportive companion offering text and voice conversations, friendly advice, and concise information in a natural, flowing style… it can provide infinite knowledge based on a person’s unique interests and needs. Pi is a teacher, coach, confidante, creative partner, and sounding board.”

“Personal AI is going to be the most transformational tool of our lifetimes. This is truly an inflection point. We’re excited to collaborate with NVIDIA, Microsoft, and CoreWeave as well as Eric, Bill and many others to bring this vision to life,” said Mustafa Suleyman, CEO and co-founder of Inflection AI.

Inflection AI describes itself as an ‘AI Studio’ specializing in creating personal AIs. It was founded in early 2022 by Suleyman, Karén Simonyan and Hoffman. The company is set up as a Public Benefit Corporation, and the Inflection AI team includes AI professionals who previously worked at DeepMind, Google, OpenAI and Meta.

“A powerful benefit of the AI revolution is the ability to use natural, conversational language to interact with supercomputers to simplify aspects of our everyday lives,” said Jensen Huang, founder and CEO of NVIDIA. “The world-class team at Inflection AI is helping to lead this groundbreaking work, deploying NVIDIA AI technology to develop, train and deploy massive generative AI models that enable amazing personal digital assistants.”

Previously, Inflection AI raised $225 million in a first round of funding in early 2022 from Greylock, Microsoft, Reid Hoffman, Bill Gates, Eric Schmidt, Mike Schroepfer, Demis Hassabis, Will.i.am, Horizons Ventures, and Dragoneer.