OpenVLM MMBench V1.1 EN - Structuralized Imagetext Understanding: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Mini (2025-08-07) | 96.3 |
| 2 | Gemini 2.5 Pro | 95.4 |
| 3 | GPT-5 | 90.8 |
| 4 | GPT-4.1 (2025-04-14) | 88.1 |
| 5 | GPT-5 Nano | 87.2 |
| 6 | GPT-4o ChatGPT | 87.2 |
| 7 | Claude 3.5 Sonnet (20241022) | 86.2 |
| 8 | Claude 3.7 Sonnet | 84.4 |
| 9 | InternVL3-38B | 82.6 |
| 10 | InternVL3-78B | 80.7 |
| 11 | InternVL2.5-78B | 80.7 |
| 12 | Claude 3.5 Sonnet | 78.9 |
| 13 | Qwen 2 VL 72B | 77.1 |
| 14 | Gemini 1.5 Pro (002) | 77.1 |
| 15 | GPT-4.5 | 75.2 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-1-en-structuralized-imagetext-understanding · How It Works · Data refreshed daily, snapshot 2026-09-05.