SciVQR - Chemistry: leaderboard
The chemistry questions of SciVQR (2026), 3,254 image-based scientific questions from competitions, entrance exams and graduate-level sources in English and Chinese, answered with zero-shot chain-of-thought. Accuracy in percent; higher is better.
Metric: Chemistry accuracy (%) with zero-shot chain-of-thought prompting on SciVQR (3,254 image-based scientific questions from competitions, entrance exams and graduate-level sources, 78 percent multiple-choice and 22 percent free-form, in English and Chinese, high-school to graduate level); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 2.5 Flash (Preview 04-17) | 87.6 | #457 | |
| 2 | O4 Mini | 85.4 | #372 | |
| 3 | O1 | 79.5 | #407 | |
| 4 | MiMo-VL-7B-RL-2508 | 74.7 | #829 | |
| 5 | Qwen 2.5 VL 72B Instruct | 66.6 | #849 | |
| 6 | GPT-4o | 60.1 | #754 | |
| 7 | Keye-VL-1.5-8B | Thinking | 53.9 | #1729 |
| 8 | InternVL3-78B-Instruct | 52 | #1436 | |
| 9 | InternVL3-14B-Instruct | 49.6 | #1810 | |
| 10 | Qwen 2.5 VL 7B Instruct | 43.2 | #1919 | |
| 11 | InternVL3-8B-Instruct | 38.4 | #2906 | |
| 12 | LLaVA-OneVision-7B | 24.6 | #3002 | |
| 13 | Molmo-7B-D-0924 | 23.4 | #7206 | |
| 14 | MiniCPM-o-2.6 | 22.7 | #2442 | |
| 15 | LLaVA-NeXT-7B | 14.1 | #5045 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=scivqr-chemistry · How It Works · Data refreshed daily, snapshot 2026-11-22.