GUI Perception and Reasoning on MPR-GUI-Bench ZH
86.2AUGemini-1.5-Flash
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Gemini-1.5-FlashModel Category=Close-source LVLMs2025.11 | 86.2 | 81.4 | 68.5 | 89.9 | 64.5 | 49.2 | 68 | 64 | 70.5 | |
| Qwen-2.5-VL-7B-InstructModel Category=Open-source LVLMs2025.11 | 83.6 | 88.8 | 77.8 | 88.8 | 79.2 | 74.3 | 68 | 68 | 80.4 | |
| Keye-VL-7BModel Category=Open-source LVLMs2025.11 | 82 | 81.4 | 73.4 | 89.3 | 77.3 | 68 | 40 | 52 | 66.9 | |
| Gemini-2.5-ProModel Category=Close-source LVLMs2025.11 | 78.4 | 85.8 | 82.5 | 81.2 | 82.5 | 71 | 92 | 84 | 82.9 | |
| MiniCPM-o 2.6Model Category=Open-source LVLMs2025.11 | 76.5 | 82 | 75.1 | 90.2 | 75.1 | 73.8 | 80 | 64 | 75.9 | |
| Intern2.5VL-8BModel Category=Open-source LVLMs2025.11 | 72.4 | 85.5 | 75.1 | 88 | 78.4 | 67.8 | 64 | 60 | 71.9 | |
| CogAgent-9BModel Category=GUI Agents2025.11 | 62.8 | 74.3 | 59.3 | 78.1 | 54 | 31.6 | 60 | 36 | 55 |