OpenVLM AI2D - Volcano — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 100 |
| 2 | InternVL3-8B | 100 |
| 3 | Qwen 2 VL 72B | 100 |
| 4 | Qwen 2 VL 7B | 100 |
| 5 | InternVL3-38B | 100 |
| 6 | Gemini 1.5 Pro (002) | 100 |
| 7 | Claude 3.7 Sonnet | 100 |
| 8 | Gemma 3 27B | 100 |
| 9 | Gemini 1.5 Pro | 100 |
| 10 | GPT-4.1 (2025-04-14) | 100 |
| 11 | GPT-4o ChatGPT | 100 |
| 12 | GPT-4.5 | 100 |
| 13 | GPT-4.1 Mini | 93.8 |
| 14 | InternVL3-78B | 93.8 |
| 15 | InternVL3-14B | 93.8 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-ai2d-volcano · How the rankings work · Data refreshed daily, snapshot 2026-07-22.