OpenVLM MMBench V1.1 CN - OCR — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 284 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 2 (1212) | 100 |
| 2 | GPT-4.1 (2025-04-14) | 100 |
| 3 | GPT-5 (2025-08-07) | 100 |
| 4 | InternVL3-78B | 98.9 |
| 5 | InternVL3-38B | 98.9 |
| 6 | GPT-4o ChatGPT | 98.9 |
| 7 | GPT-5 Mini (2025-08-07) | 98.9 |
| 8 | Qwen 2 VL 72B | 97.8 |
| 9 | Qwen 2 VL 7B | 97.8 |
| 10 | InternVL3-14B | 97.8 |
| 11 | Ovis2-8B | 97.8 |
| 12 | Gemini 1.5 Pro (002) | 97.8 |
| 13 | InternVL2.5-78B | 97.8 |
| 14 | Gemini 2.5 Pro | 96.7 |
| 15 | InternVL3-8B | 96.7 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-1-cn-ocr · How the rankings work · Data refreshed daily, snapshot 2026-07-22.