GuacaMol
Goal-directed + distribution-learning benchmarks for molecular generative models.
Composite score: 80.5
Rubric (1–5 per criterion)
rigor
4/5
coverage
4/5
maintenance
2/5
adoption
5/5
quality
4/5
accessibility
5/5
industry_relevance
4/5
Metadata
Stages
Lead ID / ADMETDevelopmental Candidate
Modalities
small-molecule
Task types
molecule-generation
License
MIT
First release
2019-03
Last updated
2022-07
Flags
none
Size & scope
- tasks: 20
- train_set: 1600000
Primary paper
Title
GuacaMol: Benchmarking Models for de Novo Molecular Design
Authors
Brown N, Fiscato M, Segler MHS, Vaucher AC
Year
2019
DOI / arXiv
Citations
820
Links
- Official site: https://www.benevolent.com/guacamol
- GitHub: https://github.com/BenevolentAI/guacamol
- Leaderboard: N/A
Hosted by (initiatives)
Experts (primary authors / maintainers)
Groups (host labs / companies / consortia)
Related benchmarks
Notes (honest caveats)
First-generation generative benchmark; largely superseded by PMO for goal-directed.