Deep Research on GAIA (test)
49.2Mean Correct RateOur best agent
Evaluation Results
| Method | Links | |
|---|---|---|
| Our best agentType=Live web2025.10 | 49.2 | |
| WebSailorType=Live web2025.10 | 34 | |
| Our base agentType=Live web2025.10 | 26.2 | |
| DeepResearcherType=Live web2025.10 | 20.63 | |
| Search-R1Type=Static & simulated web2025.10 | 18.69 | |
| ASearcherType=Live web2025.10 | 16.91 | |
| ZeroSearchType=Static & simulated web2025.10 | 8.37 | |
| R1-SearcherType=Static & simulated web2025.10 | 4.89 |