SMDD-Bench - 2D Pharmacophore Identification: leaderboard
Metric: Success rate (%) on the 25 SMDD-Bench 2D pharmacophore identification tasks, minimalist ReAct agent harness with RDKit-style tools, 8 Boltz-2 and 15 ADMET-AI oracle calls per task, temperature 1.0, at most 100 turns; success means the submitted molecule passes the task's hidden evaluator; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 28 |
| 2 | Gemini 3.1 Pro (Preview) (Medium) | 20 |
| 3 | MiniMax-M2.7 | 16 |
| 4 | Kimi K2.5 (Thinking) | 12 |
| 5 | GPT-5.4 (Medium) | 12 |
| 6 | DeepSeek V3.2 | 8 |
| 7 | Qwen 3.5 397B A17B | 4 |
Interactive version: theaggregate.ai/benchmark?slug=smdd-bench-2d-pharmacophore-identification · How It Works · Data refreshed daily, snapshot 2026-10-07.