New York – November 06, 2026 -- Inductive took first place in the OpenADMET CYP Direct Inhibition Blind Challenge, ranking ahead of more than 270 eligible participants. It is the company's fourth consecutive first-place finish in an OpenADMET competition.
NVIDIA finishes third and Recursion fourth in the CYP field
Participants predicted inhibition potency (pIC50) against four CYP isoforms (CYP3A4, CYP2D6, CYP2C9 and CYP1A2) for 750 compounds. The measurements for those compounds were withheld until the challenge closed. The field included teams from leading biopharma, academic and AI organizations.
CYP enzymes metabolize the majority of small molecule drugs. Compounds that inhibit them can cause dangerous drug-drug interactions, so drug hunters need to flag this liability early and steer designs away from it.
Inductive's Beacon models have won four blind challenges across 19 endpoints
Earlier wins came in the ASAP-Polaris-OpenADMET Antiviral Challenge (2025), the OpenADMET-ExpansionRx Challenge (February 2026) and the OpenADMET PXR Blind Challenge (July 2026). Across the four challenges, Beacon was tested on 19 drug discovery endpoints and outperformed more than 1,000 competitors.
Consortium data feeds models fine-tuned to each partner's chemical space
Beacon runs on Inductive's pre-competitive data consortium, described as one of the industry's largest and most diverse centralized ADMET datasets. The models are fine-tuned to each partner's chemical space.
Co-Founder and CTO Ben Birnbaum said data is the bottleneck for AI in chemistry, because it comes from expensive experiments that take weeks to months of expert lab work. Inductive is built around assembling the largest chemistry datasets, he said.
Co-Founder and CEO Josh Haimson said the company pairs models like Beacon with Indy, its AI chemistry agent. The aim is to help chemists make higher-quality decisions on each design, so programs need fewer design cycles to reach better development candidates.