ReactBench (Reaction Diagrams) - Localization: leaderboard
ReactBench chemical reaction diagram questions that ask which nodes are sources, sinks or intermediates from their position in the diagram topology. Exact-match accuracy in percent, higher is better. It is unrelated to the ReactBench React UI coding benchmark.
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 December 2027. 24 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 32B (Thinking) | 63.59 | #573 |
| 2 | Qwen 3 VL 32B Instruct | 60.24 | #606 |
| 3 | Gemini 3.1 Pro (Preview) | 59.16 | #99 |
| 4 | Claude Sonnet 4.6 | 55.69 | #173 |
| 5 | GPT-5.5 | 55.21 | #54 |
| 6 | Qwen 3 VL 8B Instruct | 52.34 | #1029 |
| 7 | Claude 3.5 Sonnet | 50.78 | #752 |
| 8 | Qwen 3 VL 8B (Thinking) | 49.22 | #910 |
| 9 | Qwen-VL-Max (ReactBench checkpoint unspecified) | 49.22 | |
| 10 | Qwen 2.5 VL 72B | 46.47 | #781 |
| 11 | Gemini 1.5 Pro | 45.75 | #1063 |
| 12 | DeepSeek-VL2-Tiny | 45.39 | #6534 |
| 13 | GPT-4o | 44.07 | #751 |
| 14 | InternVL2.5-78B | 41.08 | #1031 |
| 15 | InternVL2.5-8B-MPO | 38.2 | #2012 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=reactbench-reaction-diagrams-localization · How It Works · Data refreshed daily, snapshot 2026-11-21.