URSA Single-Step Retrosynthesis (USPTO-50K) - Max ChemCensor: leaderboard
Metric: Average over the targets of the authors' curated USPTO-50K test set of the highest ChemCensor score (0 to 5) among the model's predicted reactant sets (15 independent samples per target, five-shot prompts drawn from USPTO-full training reactions, no tools, reasoning enabled at the default budget for models that support it; ChemCensor 0.5.2 scores each reaction from 0 (no precedent) to 5 by the substructural context of its reaction centre and functional groups found in USPTO-full precedents); higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 3.37 | #138 (Claude Sonnet 4.5) |
| 2 | Claude Opus 4.5 (Thinking) | 3.31 | #79 (Claude Opus 4.5) |
| 3 | GPT-5.2 (Thinking) | 2.04 | #105 (GPT-5.2) |
| 4 | GPT-5.1 (Thinking) | 1.54 | #131 (GPT-5.1) |
| 5 | DeepSeek V3.2 (Thinking) | 1.14 | #198 (DeepSeek V3.2) |
| 6 | Qwen 3 14B | 0.1 | #524 |
| 7 | Qwen 3 8B | 0.04 | #667 |
| 8 | LFM2 2.6B | 0 | #1286 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=ursa-single-step-retrosynthesis-uspto-50k-max-chemcensor · How It Works · Data refreshed daily, snapshot 2026-10-11.