ReactBench (Reaction Diagrams) - Extraction: leaderboard
ReactBench chemical reaction diagram questions that extract reaction conditions, yields and other attributes while keeping them tied to the right graph element. Exact-match accuracy in percent, higher is better. It is unrelated to the ReactBench React UI coding benchmark.
Metric: Answer accuracy (%) on the topological information extraction questions (reaction conditions, yields and labels tied to graph elements), 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 February 2027. 24 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen-VL-Max (ReactBench checkpoint unspecified) | 90.1 | |
| 2 | Qwen 2.5 VL 72B | 89.86 | #782 |
| 3 | Qwen 2.5 VL 3B | 89.61 | #2664 |
| 4 | Qwen 3 VL 32B (Thinking) | 88.89 | #577 |
| 5 | GPT-5.5 | 88.41 | #56 |
| 6 | Qwen 3 VL 32B Instruct | 86.96 | #607 |
| 7 | Qwen 3 VL 8B (Thinking) | 85.99 | #924 |
| 8 | Qwen 2.5 VL 7B | 85.75 | #1878 |
| 9 | Gemini 3.1 Pro (Preview) | 85.27 | #102 |
| 10 | Qwen 3 VL 8B Instruct | 85.02 | #1043 |
| 11 | Claude 3.5 Sonnet | 81.88 | #755 |
| 12 | GPT-4o | 79.71 | #754 |
| 13 | InternVL2.5-78B | 79.47 | #1036 |
| 14 | DeepSeek-VL2-Small | 78.5 | #2523 |
| 15 | Gemini 1.5 Pro | 75.6 | #1071 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=reactbench-reaction-diagrams-extraction · How It Works · Data refreshed daily, snapshot 2026-11-22.