OpenVLM ScienceQA Test — leaderboard
OpenCompass OpenVLM test split results for ScienceQA: multimodal science questions spanning natural science, language science, and social science with diagram-grounded reasoning.
Metric: Accuracy (%). Source: huggingface.co. Status: saturated. 218 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-38B | 99.3 |
| 2 | InternVL3-78B | 98.7 |
| 3 | InternVL3-14B | 98.6 |
| 4 | InternVL3-8B | 97.9 |
| 5 | InternVL2-8B | 97.1 |
| 6 | MiniCPM-V-2.6 | 96.7 |
| 7 | Ovis2-8B | 95.1 |
| 8 | Aquila-VL-2B | 95.1 |
| 9 | GPT-4.1 (2025-04-14) | 92.8 |
| 10 | Qwen 2 VL 72B | 91.2 |
| 11 | Claude 3.7 Sonnet | 90.9 |
| 12 | Grok 2 (1212) | 90.4 |
| 13 | Claude 3.5 Sonnet | 88.9 |
| 14 | Gemma 3 27B | 87.7 |
| 15 | Pixtral-12B | 87.2 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-scienceqa-test · How the rankings work · Data refreshed daily, snapshot 2026-07-22.