Deep Research on Aggregate (BrowseComp, BrowseComp-zh, GAIA, HLE)
69.3Overall ScoreSeed1.8
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed1.82026.02 | 69.3 | |
| GPT-5-Thinking-high2026.02 | 59.1 | |
| Gemini-3-Pro2026.02 | 52.5 | |
| REDSearcherSize=30B-A3B2026.02 | 51.6 | |
| OpenAI-o32026.02 | 49.6 | |
| Tongyi DeepResearch-30BSize=30B-A3B2026.02 | 48.5 | |
| WebSailorV2-30BSize=30B-A3B2026.02 | 46 | |
| Claude-4.5-sonnet2026.02 | 41.1 |