Deep Research on FRAMES
56AccuracyTOOLSELF
Evaluation Results
| Method | Links | |
|---|---|---|
| TOOLSELFBase Model=Qwen3-14B2026.02 | 56 | |
| OAgentsBase Model=Qwen3-14B2026.02 | 48 | |
| ReSumBase Model=Qwen3-14B2026.02 | 46.5 | |
| Co-SightBase Model=Qwen3-14B2026.02 | 46 | |
| TOOLSELFBase Model=Qwen3-8B2026.02 | 44 | |
| WebSailorBase Model=Qwen3-14B2026.02 | 42 | |
| ReSumBase Model=Qwen3-8B2026.02 | 39.5 | |
| Vanilla AgentBase Model=Qwen3-14B2026.02 | 38 | |
| OWLBase Model=Qwen3-14B2026.02 | 37 | |
| OAgentsBase Model=Qwen3-8B2026.02 | 33 | |
| Co-SightBase Model=Qwen3-8B2026.02 | 32 | |
| WebSailorBase Model=Qwen3-8B2026.02 | 28 | |
| OWLBase Model=Qwen3-8B2026.02 | 23 | |
| Vanilla AgentBase Model=Qwen3-8B2026.02 | 21 |