| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Research Task | DeepResearchBench (DRB) | Accuracy (%)95.9 | 21 | |
| Search-based Question Answering | DeepResearchBench (test) | Component Score34.63 | 12 | |
| Open-ended writing | DeepresearchBench | Overall Score46.93 | 11 | |
| Agentic Writing | DeepResearchBench | Pass@149.6 | 5 | |
| Deep Research Evaluation | DeepResearchBench | WQ63.95 | 3 |