| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Writing | WritingBench | Score85.87 | 58 | |
| Writing | WritingBench v1 (test) | Average Score85.27 | 38 | |
| Instruction Following | WritingBench (Out-of-Domain) | Average Score7.9 | 23 | |
| Controllable writing | WritingBench (WB) | WB-A Score79.8 | 17 | |
| Creative Writing | WritingBench | Score6.43 | 13 | |
| Writing capability evaluation | WritingBench November 2025 (official leaderboard) | Overall Score83.87 | 9 | |
| Long-form generation | WritingBench | Score5.1 | 6 |