OpenVLM MMBench V1.0 CN - CP: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 230 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL2.5-78B | 90.6 |
| 2 | InternVL3-78B | 89.5 |
| 3 | GPT-4.1 (2025-04-14) | 89.3 |
| 4 | GPT-4.1 Mini | 87.6 |
| 5 | Claude 3.7 Sonnet | 87.6 |
| 6 | Qwen 2 VL 72B | 87.4 |
| 7 | Qwen 2 VL 7B | 86.9 |
| 8 | InternVL2-8B | 86.9 |
| 9 | InternVL3-8B | 86.1 |
| 10 | InternVL3-14B | 86.1 |
| 11 | Gemma 3 27B | 85.9 |
| 12 | Gemini 1.5 Flash | 85.7 |
| 13 | Claude 3.5 Sonnet | 85.4 |
| 14 | Claude 3.5 Sonnet (20241022) | 84.6 |
| 15 | Grok 2 (1212) | 83.9 |
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-0-cn-cp · How It Works · Data refreshed daily, snapshot 2026-09-19.