| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| OSWorld | UI-TARS | Average Accuracy42.5 | 24 | 1mo ago | |
| Memory-World (test) | Stamp-GUI | T-Acc82.6 | 22 | 5d ago | |
| OSWorld | Success Rate (Max Steps: 15)42.9 | 16 | 5d ago | ||
| MobileWorld GUI-Only | SR55.6 | 14 | 1mo ago | ||
| OSWorld w/o Loop | GPT-4o + ScaleCUA-7B | AUV29.5 | 5 | 3mo ago | |
| OSWorld w/ Loop | AUV6.9 | 5 | 3mo ago |
