ReactBench (Reaction Diagrams) - Localization: leaderboard
Metric: Answer accuracy (%) on the spatial element localization questions (identify source, sink and intermediate nodes from topological position), over the 1,618 expert-annotated question-answer pairs on more than 1,300 real chemical reaction diagrams from chemistry journals and patents of ReactBench (single-line, multi-line, tree and graph topologies), multiple-choice or short open answers matched exactly against the key after template extraction, no LLM judge; higher is better. Source: arxiv.org. Saturation forecast: Around September 2028. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 32B (Thinking) | 63.59 |
| 2 | Qwen 3 VL 32B Instruct | 60.24 |
| 3 | Gemini 3.1 Pro (Preview) | 59.16 |
| 4 | Claude Sonnet 4.6 | 55.69 |
| 5 | GPT-5.5 | 55.21 |
| 6 | Qwen 3 VL 8B Instruct | 52.34 |
| 7 | Claude 3.5 Sonnet | 50.78 |
| 8 | Qwen 3 VL 8B (Thinking) | 49.22 |
| 9 | Gemini 1.5 Pro | 45.75 |
| 10 | GPT-4o | 44.07 |
| 11 | InternVL2.5-78B | 41.08 |
| 12 | Qwen 2.5 VL 7B | 37.96 |
Interactive version: theaggregate.ai/benchmark?slug=reactbench-reaction-diagrams-localization · How It Works · Data refreshed daily, snapshot 2026-10-07.