MolDesignBench - Molecule Optimization: leaderboard
The optimization task of MolDesignBench (Jeong et al., 2026): the agent must modify a given starting molecule so that it satisfies every implicit and explicit constraint while staying structurally similar, or answer None when the request is infeasible, using a 17-tool chemistry toolset through function calling. Success rate in percent over 1,000 instances, one tenth of them infeasible; higher is better.
Metric: Success rate (%; share of the 1,000 optimization instances handled correctly: a valid SMILES satisfying every implicit and explicit property and functional-group constraint on the 900 feasible instances, staying similar to the given starting molecule, or an explicit None on the 100 infeasible ones; agent with the 17-tool chemistry toolset through function calling, at most 20 turns). Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 43 | #121 |
| 2 | Claude Opus 4.6 (Thinking) | 41 | #100 (Claude Opus 4.6) |
| 3 | Qwen 3 235B A22B (Thinking) | 25 | #622 (Qwen 3 235B A22B) |
| 4 | GPT-OSS-120B | 23 | #699 |
| 5 | Intern-S1 | 10 | #616 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=moldesignbench-molecule-optimization · How It Works · Data refreshed daily, snapshot 2026-11-12.