Lilly opens its AI drug discovery models to outside biotechs with TuneLab

On Sept. 9, 2025, Eli Lilly and Company launched Lilly TuneLab, a platform that gives biotechnology companies access to artificial intelligence and machine learning models trained on years of Lilly's drug discovery data. Lilly estimates the first release of models includes proprietary data obtained at a cost of over $1 billion, which the company describes as one of the industry's most valuable datasets used to train an AI system available to biotech companies ^[1].

TuneLab is built on what Lilly calls its full drug disposition, safety and preclinical datasets, described as experimental data obtained with hundreds of thousands of unique molecules ^[1]. In return for access, selected biotech partners contribute training data, which Lilly says fuels continuous improvement of the models for others in the ecosystem ^[1]. The platform is hosted by a third party and uses federated learning, an approach Lilly describes as privacy-preserving: biotechs can tap into Lilly's models without directly exposing their own proprietary data, and Lilly says its data is not directly exposed either ^[1].

"Lilly TuneLab was created to be an equalizer so that smaller companies can access some of the same AI capabilities used every day by Lilly scientists," said Daniel Skovronsky, M.D., Ph.D., Lilly's chief scientific officer and president of Lilly Research Laboratories and Lilly Immunology ^[1].

Nisha Nanda, Ph.D., group vice president and head of Lilly Catalyze360, said most small biotechs lack the large-scale, high-quality data needed to train effective models. "With Lilly TuneLab we're not just sharing resources, we are also compressing decades of learning into instantly accessible intelligence," she said ^[1].

TuneLab is the newest addition to Lilly Catalyze360, the company's set of offerings for biotech partners, which also includes strategic capital through Lilly Ventures, laboratory facilities at Lilly Gateway Labs, and drug development expertise through Lilly ExploR&D ^[1]. Lilly says it intends to extend the platform's features beyond this first release, including adding in vivo small molecule predictive models that would be available exclusively on Lilly TuneLab. No date was given for that expansion ^[1].

The announcement does not identify which diseases, drug classes or specific medicines the models are meant to support, how much partners pay for access, how many partners have signed on, or how long it takes to be selected. The release also does not mention semaglutide, tirzepatide or any specific GLP-1 or obesity medicine, and it describes no change to medicines already on the market ^[1].

Why it matters for patients

TuneLab is aimed at early-stage drug discovery, the point where companies decide which molecules to pursue. Medicines that come out of that stage typically take years to reach patients, if they reach them at all, and Lilly does not claim that any specific medicine will result from the platform ^[1]. The company says it hopes to "increase their likelihood of success" for biotech partners, but the release reports no evidence yet that the models do so ^[1].

For anyone taking a GLP-1 medication today, nothing in this announcement changes what is available, what it costs or how it is prescribed. The practical effect patients could eventually see would come indirectly, if faster or better early research adds new products to Lilly's pipeline down the road. That has not happened yet, and no timeline for it has been disclosed ^[1].

What happens next

Lilly says it plans to add in vivo small molecule predictive models to TuneLab, available only on the platform, but has not said when ^[1]. The company directs interested biotechs to its TuneLab site for partnership information; no application deadlines or selection criteria were disclosed ^[1]. Lilly also notes in a standard cautionary statement that there is no assurance the platform or Catalyze360 will achieve the company's objectives or that Lilly will execute its strategy as planned ^[1].