MoleculeNet
The original molecular-machine-learning benchmark of 17 dataset collections and more than 800 prediction endpoints spanning quantum, physicochemical, biophysical, and physiological properties.
Scientific task · Design and generation
Small-molecule discovery
Generate
小分子发现
Definition and search aliases
- Permanent ID
small-molecule-discovery- Aliases
- drug discovery
- Deprecated aliases
- None
- Hierarchy
- Parent task; benchmark coverage below includes its leaf tasks.
Coverage
3 benchmark families cover this task
GuacaMol
A reproducible benchmark for de novo molecular design with five distribution-learning tests and twenty goal-directed generation problems in the current v2 suite.
ATOM3D
A living collection of eight curated 3D molecular-learning tasks spanning small molecules, protein interactions and mutations, ligand binding, and protein/RNA structure ranking.
Evidence-backed count claims
Each row keeps its original unit and basis. Rows with different units or overlapping mappings are never added.
| Benchmark | Mapped task | Coverage | Count | Version | Evidence |
|---|---|---|---|---|---|
| MoleculeNet root: moleculenet | ADMET and toxicity prediction official-taxonomy · high | explicitly-in-scope | 5 other Original paper dataset collections. | original-2017 | moleculenet-paper-definition-evidence |
| GuacaMol root: guacamol | Small-molecule generation official-taxonomy · high | explicitly-in-scope | 25 problems Formal benchmark problems. | suite-v2 | guacamol-paper-definition-evidenceguacamol-repository-evidence |
| ATOM3D root: atom3d | Small-molecule property prediction official-track · high | explicitly-in-scope | 1 tasks Curated 3D benchmark datasets. | v0.2.6 as of 2026-07-22 | atom3d-paper-definition-evidence |
| MoleculeNet root: moleculenet | Small-molecule property prediction official-taxonomy · high | explicitly-in-scope | 17 other Original paper dataset collections. | original-2017 | moleculenet-paper-definition-evidence |
Official evaluations connected to these benchmarks
Runs are included only for benchmark records mapped here (and formal child tracks when a mapped suite is the root). A task mapping does not imply that every run isolates this task.
| Work | Provider / class | Related runs |
|---|---|---|
| ATOM3D: Tasks On Molecules in Three Dimensions | Stanford University benchmark_creator | atom3d-creator-full |
| GuacaMol: Benchmarking Models for de Novo Molecular Design | BenevolentAI benchmark_creator | guacamol-creator-full |
| MoleculeNet: a benchmark for molecular machine learning | Stanford University, DeepChem benchmark_creator | moleculenet-creator-full |