insitro's AI drug-prediction models go live for biotechs on Lilly's TuneLab platform
What's the deal? On 6 October 2026, insitro announced that the machine-learning models it built with Eli Lilly to predict how small-molecule drugs behave in the body are now available to biotech companies using Lilly's TuneLab platform. TuneLab is Lilly's AI drug-discovery service that lets outside biotechs use shared predictive models without surrendering their proprietary data.
Why it matters: Most experimental medicines fail in clinical trials, often because the molecule never reaches the right tissue at the right concentration for the right time. Traditionally, researchers only learned how a drug behaved in a living system through slow, costly lab and animal studies. These models, trained on Lilly's decades of preclinical data from hundreds of thousands of molecules, let scientists predict a compound's behaviour computationally before it is even synthesised. The aim is shorter, cheaper paths for promising drugs to reach patients and less reliance on animal testing, which aligns with the FDA's recent push to modernise preclinical safety testing. The models also feed insitro's own TherML platform for designing medicines against the genetic drivers of disease.
Context: The collaboration builds on the insitro-Lilly partnership announced in September 2025. insitro, founded by AI pioneer Daphne Koller, is one of the most prominent AI-drug-discovery companies, backed by more than US$750M and working with Lilly, Bristol Myers Squibb and Gilead. It plans to take its first wholly owned programmes into the clinic in early 2027.
Read more: Business Wire, 6 Oct 2026
