OpenVLM MMBench V1.1 CN - Image Quality: leaderboard
OpenCompass OpenVLM MMBench V1.1 Chinese split accuracy for image quality assessment.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around January 2027. 284 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Mini (2025-08-07) | 74.2 | #172 |
| 2 | GPT-5 | 70.2 | #98 |
| 3 | GPT-4o ChatGPT | 70.2 | |
| 4 | Gemini 2.5 Pro | 68.5 | #152 |
| 5 | GPT-4.1 (2025-04-14) | 67.7 | #250 |
| 6 | InternVL3-78B | 64.5 | #352 |
| 7 | Qwen 2 VL 72B | 64.5 | #460 |
| 8 | Qwen 2.5 VL 7B | 64.5 | #643 |
| 9 | Claude 3.5 Sonnet (20241022) | 62.9 | #291 |
| 10 | InternVL2.5-78B | 62.9 | #431 |
| 11 | GPT-5 Nano | 62.1 | #425 |
| 12 | InternVL3-38B | 62.1 | #405 |
| 13 | InternVL2-8B | 61.3 | #817 |
| 14 | InternVL3-8B | 60.5 | #620 |
| 15 | Qwen 2 VL 7B | 60.5 | #750 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmbench-v1-1-cn-image-quality · How It Works · Data refreshed daily, snapshot 2026-10-20.