GUI Visual Grounding on ScreenSpot Standard Setting 1.0
94.1Accuracy (Mobile, Text)UGround-V1-72B
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| UGround-V1-72BModel Size=72B, Base Model=Qwen2-VL2024.10 | 94.1 | 83.4 | 94.9 | 85.7 | 90.4 | 87.9 | 89.4 | |
| UGround-V1-7BModel Size=7B, Base Model=Qwen2-VL2024.10 | 93 | 79.9 | 93.8 | 76.4 | 90.9 | 84 | 86.3 | |
| Aria-UIInput Resolution=3920x29402024.12 | 92.3 | 73.8 | 93.3 | 64.3 | 86.5 | 76.2 | 82.4 | |
| UGround-V1-2BModel Size=2B, Base Model=Qwen2-VL2024.10 | 89.4 | 72 | 88.7 | 65.7 | 81.3 | 68.9 | 77.7 | |
| UGround2024.10 | 82.8 | 60.3 | 82.5 | 63.6 | 80.4 | 70.4 | 73.3 | |
| UGround2024.12 | 82.8 | 60.3 | 82.5 | 63.6 | 80.4 | 70.4 | 74.1 | |
| SeeClick2024.10 | 78 | 52 | 72.2 | 30 | 55.7 | 32.5 | 53.4 | |
| SeeClick2024.12 | 78 | 52 | 72.2 | 30 | 55.7 | 32.5 | 55.8 | |
| Qwen2-VL2024.12 | 75.5 | 60.7 | 76.3 | 54.3 | 35.2 | 25.7 | 55.3 | |
| CogAgent2024.10 | 67 | 24 | 74.2 | 20 | 70.4 | 28.6 | 47.4 | |
| CogAgent2024.12 | 67 | 24 | 74.2 | 20 | 70.4 | 28.6 | 49.6 | |
| GPT-42024.10 | 22.6 | 24.5 | 20.2 | 11.8 | 9.2 | 8.8 | 16.2 | |
| GPT-42024.12 | 22.6 | 24.5 | 20.2 | 11.8 | 9.2 | 8.8 | 16.7 | |
| GPT-4o2024.10 | 20.2 | 24.9 | 21.1 | 23.6 | 12.2 | 7.8 | 18.3 | |
| GPT-4o2024.12 | 20.2 | 24.9 | 21.1 | 23.6 | 12.2 | 7.8 | 18.1 |