Lilly launches TuneLab, opening its AI drug discovery models to outside biotechs

Eli Lilly announced on September 9, 2025, that it has launched Lilly TuneLab, a platform that lets outside biotech companies use artificial intelligence and machine learning models trained on Lilly's own drug discovery data ^[1]. Lilly says the first version of these models draws on proprietary data that cost the company more than $1 billion to obtain, built from experiments on hundreds of thousands of unique molecules ^[1].

The models cover Lilly's drug disposition, safety, and preclinical research, according to the company ^[1]. Biotech partners that use the platform contribute their own training data in return, which Lilly says helps improve the models over time for the benefit of the wider group of users ^[1]. To protect proprietary information, the platform is hosted by a third party and uses a technique called federated learning, which Lilly describes as a privacy-preserving method that lets biotechs tap into the models without directly exposing their own data or Lilly's ^[1].

Daniel Skovronsky, Lilly's chief scientific officer and president of Lilly Research Laboratories and Lilly Immunology, said the goal is to give smaller companies access to "some of the same AI capabilities used every day by Lilly scientists" ^[1]. Nisha Nanda, who heads Lilly's Catalyze360 program for biotech partners, said many early-stage biotechs lack the large-scale, high-quality data needed to train effective AI models on their own, and that TuneLab is meant to close that gap ^[1].

TuneLab is the newest piece of Lilly Catalyze360, an existing set of programs Lilly offers to biotech partners, which also includes venture capital funding through Lilly Ventures, lab space through Lilly Gateway Labs, and drug development guidance through Lilly ExploR&D ^[1]. Lilly says it plans to keep expanding TuneLab's features, including adding models that predict how small-molecule drugs behave inside living organisms, which the company says will be available only through this platform ^[1].

Why it matters for patients

TuneLab is a research tool aimed at biotech companies and scientists, not a product patients will use directly, and it does not change how any current GLP-1 medicine like Ozempic, Wegovy, Mounjaro, or Zepbound works or is prescribed. The practical relevance for patients is indirect: Lilly is betting that giving smaller biotech firms access to its AI models, built from more than a billion dollars of research spending, could help those companies make faster, more informed decisions earlier in drug development ^[1]. Faster and cheaper early-stage research does not guarantee that new treatments reach patients sooner, since drugs still need to go through clinical trials and regulatory review before they can be prescribed. The announcement gives no timeline for when, or whether, any specific new medicine will result from work done on the platform.

What happens next

Lilly has not announced specific drug candidates coming out of TuneLab, and the company describes this launch as a first release, with plans to add more predictive models later ^[1]. Companies interested in partnering with Lilly can apply through the platform's website, but the sources do not name which biotech firms, if any, have already signed on ^[1]. As with any forward-looking corporate announcement, Lilly cautions in its own release that there is no guarantee the program will meet its stated goals ^[1].