TDC ADMET Group

22-task ADMET benchmark suite with scaffold splits — core TDC leaderboard.

Composite

100.0

Experimental validation

Retrospective

Stages

Lead ID / ADMET

Modalities

small-molecule

Task types

regressionclassification

Size

tasks: 22
molecules: 130,000

License

MIT

First release

2021-02

Last updated

2025-04

Official site

→ project page

Leaderboard

→ leaderboard

Dataset

→ dataset

Code / GitHub

→ repository

HuggingFace

→ HF

Paper

Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Huang K, Fu T, Gao W, et al. · 2021 · paper · doi:10.48550/arXiv.2102.09548 · 620 citations

Flags

none

Experts

Kexin Huang, Marinka Zitnik, Tianfan Fu

Groups

Zitnik Lab, Therapeutics Data Commons

Hosted by

Therapeutics Data Commons (TDC), ScienceAIBench, Drug Discovery Benchmarks (DDB)

Related benchmarks

MoleculeNet, ADMET-AI, Polaris ADMET, MoleculeACE

Rubric (7-criterion)

rigor

coverage

maintenance

adoption

quality

accessibility

industry_relevance

Notes

Most-adopted ADMET benchmark. 100+ leaderboard submissions.

← Back to all benchmarks

Compare:

Open comparison →