A pharmaceutical company just switched on a supercomputer that can test billions of potential drug molecules before a single lab experiment is run.
Published in Artificial Intelligence, Drugs development & discovery, Supercomputing/Calculations.
A pharmaceutical company just switched on a supercomputer that can test billions of potential drug molecules before a single lab experiment is run.
Eli Lilly inaugurated LillyPod at its Indianapolis headquarters in February 2026, confirmed by the NVIDIA Blog, R&D World, and Lilly’s own investor communications. Built on NVIDIA’s DGX SuperPOD architecture powered by 1,016 Blackwell Ultra GPUs, LillyPod delivers over 9,000 petaflops of AI performance, the equivalent of nine quintillion mathematical operations every second, and was fully assembled in just four months. Traditional pharmaceutical wet labs can physically evaluate roughly 2,000 molecular hypotheses per drug target per year. LillyPod eliminates that physical ceiling entirely: “Now in the dry lab, you can test billions of molecule ideas at your fingertips,” said Yue Wang Webster, Lilly’s VP of research and development informatics. The platform runs workloads across genomics, molecule design, protein modeling, single-cell biology, medical imaging, and manufacturing optimization simultaneously.
Lilly also built a federated AI platform called TuneLab on top of LillyPod, trained on over one billion dollars worth of internal research data, including lessons learned from millions of failed drug candidates, and is opening access to smaller biotech companies so they can run Lilly’s predictive models within their own existing workflows.
–
News Source: NVIDIA Blog / Eli Lilly official, “Now Live: Lilly AI Factory for Pharmaceutical Discovery and Development” (February 27, 2026)
–
#EliLilly #AIInMedicine #DrugDiscovery #Pharma #NVIDIA
