| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Safety | T3 | T3 Score85.1 | 21 | |
| Stitched image rectangling | T3 (test) | PSNR25.1 | 4 | |
| Research Assistant | T3 Research 1.0 (test) | Task Completion Rate88 | 4 | |
| Task T3 | T3 | Token Usage (Input + Output)2,156 | 4 | |
| Predictive Modeling | T3 | Loss0.063 | 3 |