Eli Lilly, Nvidia to build supercomputer, AI factory for drug discovery

Eli Lilly And Nvidia They are partnering to build what they call the pharmaceutical industry’s “most powerful” supercomputer and a so-called artificial intelligence factory to help accelerate drug discovery and development across the industry, the companies said Tuesday.

This is the latest step by Nvidia and the pharmaceutical industry to leverage AI to help shorten the time it takes to deliver treatments to patients, while also reducing costs at every stage of drug discovery and development. The process typically takes an average of 10 years from the first human drug to launch, Diogo Rau, Eli Lilly’s chief information officer, said in an interview.

Eli Lilly expects to complete construction of its supercomputer and artificial intelligence factory in December. They will be online in January. But the new tools likely won’t deliver significant returns to the company’s business and that of other drugmakers until the end of the decade.

“With this kind of power that we have now, we will actually see the benefits of the things we are talking about exploring in 2030,” Rau said.

The industry’s efforts to use artificial intelligence to get medicines to people faster are still in their early stages. There are no drugs on the market designed using AI, but progress is clearly evident Number of drugs discovered by AI entering clinical trials, more recently AI-focused investments, and partnerships between pharmaceutical manufacturers.

Eli Lilly will own and operate the supercomputer, which will be powered by more than 1,000 Blackwell Ultra GPUs (a newer chip family from Nvidia) connected to a unified, high-speed network. The supercomputer will power the AI ​​factory, a specialized computing infrastructure that will develop, train and deploy AI models at scale for drug discovery and development.

The supercomputer is “truly a new scientific tool. It’s like a giant microscope for biologists,” said Thomas Fuchs, Eli Lilly’s chief AI officer. “It really allows us to do things on such a large scale that we couldn’t do before.

According to a statement from Eli Lilly, scientists will be able to train AI models on millions of experiments to test potential drugs, thereby “significantly expanding the scope and complexity” of drug discovery.

Finding new drugs That’s not the sole focus of new vehicles, but “where the big opportunity is,” Rau said.

“We hope that we can discover new molecules that we could never discover with humans alone,” he said.

A variety of AI models will be available on Lilly TuneLab, an AI and machine learning platform that allows biotech companies to access drug discovery models that Eli Lilly has trained over years of proprietary research. The value of this data is 1 billion dollars.

Eli Lilly launched the platform in September as a way to expand access to drug discovery tools across the industry.

“It’s really powerful to be able to give an extra starting point to these startups that might otherwise have taken a few years to burn capital to get to this point,” said Kimberly Powell, Nvidia’s vice president of healthcare, adding that the company was “happy to join” the effort.

In exchange for access to artificial intelligence models, biotech companies are expected to contribute some of their own research and data to help train them, Rau said. The TuneLab platform uses technology called federated learning; This means companies can leverage Lilly’s AI models without requiring either party to directly share data.

Eli Lilly also plans to use the supercomputer to shorten the drug development process and help get treatments to people faster.

The company said new scientific AI agents could support researchers and advanced medical imaging could give scientists a clearer view of how diseases progress and help them develop new biomarkers (a measurable sign of a biological process or condition) for personalized care.

“We actually want to deliver on this promise of precision medicine,” Powell said. “We’ll never get there without the AI ​​infrastructure and foundation, right? So we’ve been doing all the necessary building, and now we’re seeing this real takeoff, and Lilly is a perfect example of that.”

Precision medicine is an approach that tailors disease prevention and treatment to differences in a person’s genes, environment, and lifestyle.