Benchmark
MMBench
vision
multimodal
reasoning
multimodal
多语言
A bilingual benchmark for assessing multi-modal capabilities of vision-language models through multiple-choice questions in both English and Chinese, providing systematic evaluation across diverse vision-language tasks with robust metrics.
语言EN
满分1
参评模型7
模型排名
| 名次 | 模型 | 机构 | 分数 | 来源 |
|---|---|---|---|---|
| 1 | Qwen2.5 VL 72B Instruct | Alibaba Cloud / Qwen Team | 88.0 | 来源 ↗ |
| 2 | Phi-4-multimodal-instruct | Microsoft | 86.7 | 来源 ↗ |
| 3 | Qwen2.5 VL 7B Instruct | Alibaba Cloud / Qwen Team | 84.3 | 来源 ↗ |
| 4 | Phi-3.5-vision-instruct | Microsoft | 81.9 | 来源 ↗ |
| 5 | DeepSeek VL2 Small | DeepSeek | 80.3 | 来源 ↗ |
| 6 | DeepSeek VL2 | DeepSeek | 79.6 | 来源 ↗ |
| 7 | DeepSeek VL2 Tiny | DeepSeek | 69.2 | 来源 ↗ |