GUI Perception and Reasoning on MPR-GUI-Bench JA
0.852AUGemini-2.5-Pro
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Gemini-2.5-ProModel Category=Close-source LVLMs2025.11 | 0.852 | 0.847 | 0.534 | 0.65 | 0.601 | 0.623 | 0.68 | 0.76 | 0.7 | |
| Qwen-2.5-VL-7B-InstructModel Category=Open-source LVLMs2025.11 | 0.817 | 0.877 | 0.792 | 0.907 | 0.773 | 0.691 | 0.88 | 0.68 | 0.795 | |
| Gemini-1.5-FlashModel Category=Close-source LVLMs2025.11 | 0.803 | 0.828 | 0.712 | 0.888 | 0.52 | 0.447 | 0.64 | 0.4 | 0.627 | |
| Keye-VL-7BModel Category=Open-source LVLMs2025.11 | 0.781 | 0.822 | 0.712 | 0.866 | 0.727 | 0.582 | 0.36 | 0.4 | 0.614 | |
| MiniCPM-o 2.6Model Category=Open-source LVLMs2025.11 | 0.727 | 0.776 | 0.74 | 0.858 | 0.74 | 0.672 | 0.68 | 0.64 | 0.717 | |
| Intern2.5VL-8BModel Category=Open-source LVLMs2025.11 | 0.642 | 0.828 | 0.729 | 0.806 | 0.732 | 0.691 | 0.64 | 0.44 | 0.66 | |
| CogAgent-9BModel Category=GUI Agents2025.11 | 0.483 | 0.735 | 0.548 | 0.68 | 0.36 | 0.263 | 0.4 | 0.44 | 0.479 |