| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| GUI-Odyssey | OS-Atlas | Type Match (TM)91.83 | 20 | 22d ago | |
| Offline Benchmark | Exact Match69.2 | 15 | 1mo ago | ||
| ScreenDrag offline (test) | OpenCUA-7B | Endpoint Acc (OS)99.01 | 10 | 3mo ago | |
| Meta-GUI (test) | OS-Kairos | Scroll Rate99.36 | 7 | 3mo ago | |
| Android Control High | Mobile-R1 | Task Match (TM)76.5 | 6 | 1mo ago | |
| Android Control Low | Qwen2.5-VL | TM94.1 | 6 | 1mo ago | |
| AITZ | Mobile-R1 | Task Match (TM)77.05 | 5 | 1mo ago |