Inductive Wins OpenADMET CYP Challenge for Fourth Straight Time
Inductive placed first in the OpenADMET CYP Direct Inhibition Blind Challenge, finishing ahead of more than 270 participants from biopharma, academia and AI organizations. The result gives the chemical intelligence company its fourth consecutive first-place finish in OpenADMET blind drug discovery competitions.
- Inductive ranked first in the OpenADMET CYP Direct Inhibition Blind Challenge.
- The challenge included more than 270 eligible participants and 750 compounds.
- The models predicted pIC50 for CYP3A4, CYP2D6, CYP2C9 and CYP1A2.
- NVIDIA placed third and Recursion placed fourth.
- The result is Inductive's fourth consecutive first-place OpenADMET finish.
A blind test of drug safety predictions
The challenge required participants to predict inhibition potency, measured as pIC50, for four major cytochrome P450 isoforms: CYP3A4, CYP2D6, CYP2C9 and CYP1A2. The measurements for 750 compounds were withheld until the competition closed.
Cytochrome P450 enzymes metabolize the majority of small molecule drugs. Compounds that inhibit these enzymes can cause dangerous drug-drug interactions, making early identification an important part of drug discovery. Predictive models can help drug hunters detect this liability and steer chemical designs away from it.
Because the compounds were previously unseen by participants, the blind format provided a test of ADMET modeling under conditions that Inductive described as closer to real-world prediction. The Direct Inhibition track attracted more than 270 eligible participants.
Inductive leads teams from NVIDIA and Recursion
Inductive's Beacon models ranked first in a field that included teams from leading AI and biopharma organizations. NVIDIA placed third and Recursion placed fourth.
The company said this is its fourth consecutive first-place OpenADMET result. Its previous wins came in the ASAP-Polaris-OpenADMET Antiviral Challenge in 2025, the OpenADMET-ExpansionRx Challenge in February 2026, and the OpenADMET PXR Blind Challenge in July 2026.
Across the four challenges, Inductive said its Beacon models have been tested on 19 critical drug discovery endpoints and have outperformed more than 1,000 competitors, including large biopharma and AI organizations.
Inductive links performance to chemistry data
Inductive said its Beacon models are powered by a pre-competitive data consortium, which the company described as one of the industry's largest and most diverse centralized ADMET datasets. The models are also fine-tuned to each partner's chemical space.
“For AI in chemistry to match the progress we've seen in coding and math, data is the bottleneck,” said Ben Birnbaum, Inductive's Co-Founder and CTO. He said chemistry data comes from real experiments that are expensive and require weeks to months of expert work in chemistry labs.
Inductive also positions Beacon alongside Indy, its AI chemistry agent. “Pairing state-of-the-art models like these with Indy, our AI chemistry agent, helps each chemist make higher-quality decisions on every design, so programs need fewer design cycles to reach better development candidates,” said Josh Haimson, Inductive's Co-Founder and CEO.
For drug discovery teams, the result adds another external benchmark for Inductive's prediction models, while the company's account ties that performance to its data consortium and partner-specific model tuning. Inductive said its models, lab robotics and expert chemists support discovery programs with leading biopharma partners.
