OpenVLM AI2D - Rock Strata: leaderboard
OpenCompass OpenVLM AI2D subcategory score for diagram questions about rock strata.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 284 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3-78B | 95.1 | #352 |
| 2 | Qwen 2 VL 72B | 92.7 | #460 |
| 3 | GPT-5 Mini (2025-08-07) | 92.7 | #172 |
| 4 | GPT-4.5 | 92.7 | #274 |
| 5 | Gemini 2.5 Pro | 90.2 | #152 |
| 6 | GPT-5 | 90.2 | #98 |
| 7 | InternVL2.5-78B | 90.2 | #431 |
| 8 | InternVL3-38B | 87.8 | #405 |
| 9 | GPT-4.1 (2025-04-14) | 87.8 | #250 |
| 10 | GPT-4o ChatGPT | 87.8 | |
| 11 | GPT-5 Nano | 85.4 | #425 |
| 12 | InternVL3-14B | 85.4 | #505 |
| 13 | Qwen 2.5 VL 7B | 85.4 | #643 |
| 14 | Claude 3.7 Sonnet | 82.9 | #249 |
| 15 | Gemini 1.0 Pro | 82.9 | #933 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-ai2d-rock-strata · How It Works · Data refreshed daily, snapshot 2026-10-20.