| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| ScreenSpot Pro | MAI-UI-32B | Average Score7,350 | 482 | 25d ago | |
| ScreenSpot v2 | MAI-UI-32B | Avg Accuracy96.5 | 447 | 17d ago | |
| ScreenSpot Pro | Accuracy72.7 | 221 | 25d ago | ||
| ScreenSpot | GUI-G2 | Avg Acc92 | 169 | 1mo ago | |
| OSWorld-G | Qwen3VL-8B-Instruct | Average Score72.7 | 164 | 24d ago | |
| MMBench-GUI L2 (test) | GPT-4o | Average Error2.9 | 87 | 24d ago | |
| ScreenSpot Web V2 | UI-Venus-1.5-8B | Text Accuracy98.3 | 78 | 24d ago | |
| ScreenSpot Desktop V2 | POINTS-GUI-G-8B | Text Accuracy100 | 78 | 24d ago | |
| ScreenSpot Mobile V2 | UI-AGILE-7B | Text Accuracy100 | 78 | 24d ago | |
| UI-Vision | GroundNext-3B (RL) | Average Score62.1 | 68 | 1mo ago | |
| ScreenSpot-Pro (test) | VISTA | Element Accuracy74.1 | 67 | 1mo ago | |
| MMBench-GUI-L2 | MAI-UI-32B | Accuracy91.3 | 63 | 24d ago | |
| UI-Vision (test) | MAI-UI-32B | Basic Score59.1 | 59 | 1mo ago | |
| OSWorld-G (test) | GTA1-32B | Element Accuracy78.4 | 52 | 4mo ago | |
| ScreenSpot (test) | Trifuse+GUI-AIMA | Element Accuracy90.6 | 42 | 3mo ago | |
| Screenspot v1 | Qwen3-VL-32B | Average Accuracy91.67 | 39 | 3mo ago | |
| UI-Vision | GROUNDNEXT-7B (SFT) | Accuracy58.7 | 38 | 25d ago | |
| OSWorld G-Refine v1.0 (test) | Overall Success Rate76.1 | 37 | 1mo ago | ||
| ScreenSpot v2 (test) | VISTA | Element Accuracy95.6 | 33 | 1mo ago | |
| MMBench GUI | Accuracy88.62 | 32 | 25d ago | ||
| ScreenSpot Pro | UI-Venus-72B | Click Accuracy61.9 | 25 | 1mo ago | |
| ScreenSpot v1 (test) | GUI-G1 | Mobile Text Acc98.6 | 25 | 4mo ago | |
| OSWorld-G-Refine | UI-Venus-1.5-30B-A3B | Accuracy76.4 | 24 | 24d ago | |
| OSWorld-G v1.0 (test) | VISTA | Accuracy67.6 | 22 | 1mo ago | |
| ScreenSpot V1 (Overall) | UGround-V1-7B | Average Accuracy89.9 | 22 | 2mo ago |