Comparative Performance Evaluation on DeepConsult
77.21Win RateADORE
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ADORE2026.01 | 77.21 | 7.03 | 18.38 | 4.41 | |
| Enterprise Deep Research2026.01 | 71.57 | 6.82 | 19.12 | 9.31 | |
| WebWeaverBackbone=Claude-sonnet2026.01 | 66.86 | 6.96 | 10.47 | 22.67 | |
| WebWeaverBackbone=gpt-oss-120b2026.01 | 65.31 | 6.64 | 11.22 | 23.47 | |
| Gemini-2.5-pro2026.01 | 61.27 | 6.7 | 31.13 | 7.6 | |
| WebWeaverBackbone=qwen3-235b2026.01 | 54.74 | 6.47 | 28.61 | 16.67 | |
| Perplexity Deep Research2026.01 | 32 | — | — | — | |
| doubao-research2026.01 | 29.95 | 5.42 | 40.35 | 29.7 | |
| WebWeaverBackbone=qwen3-30b2026.01 | 28.65 | 4.57 | 34.9 | 36.46 | |
| Claude-research2026.01 | 25 | 4.6 | 38.89 | 36.11 | |
| WebShaperParameters=32B2026.01 | 3.25 | 1.63 | 3.75 | 93 | |
| Enterprise DRSystem category=Prompt-Based Frameworks, Base model=Gemini-2.5-Pro2026.02 | 0.7157 | 6.82 | 0.1912 | 0.0931 | |
| RhinoInsighSystem category=Prompt-Based Frameworks, Base model=Gemini-2.5-Pro2026.02 | 0.6851 | 6.82 | 0.1102 | 0.2047 | |
| WebWeaverSystem category=Prompt-Based Frameworks, Base model=Claude-Sonnet-42026.02 | 0.6686 | 6.96 | 0.1047 | 0.2267 | |
| Gemini-2.5-Pro-deepresearchSystem category=Proprietary Deep Research Systems2026.02 | 0.6127 | 6.7 | 0.3113 | 0.076 | |
| AgentCPM-Report (Pipeline RL)System category=Our Deep Research Systems, Training strategy=Pipeline RL2026.02 | 0.576 | 6.6 | 0.1373 | 0.2868 | |
| AgentCPM-Report (Atomic RL)System category=Our Deep Research Systems, Training strategy=Atomic RL2026.02 | 0.5613 | 6.06 | 0.1103 | 0.3284 | |
| AgentCPM-Report (SFT)System category=Our Deep Research Systems, Training strategy=SFT2026.02 | 0.5417 | 6.04 | 0.1029 | 0.3554 | |
| Doubao-researchSystem category=Proprietary Deep Research Systems2026.02 | 0.2995 | 5.42 | 0.4035 | 0.297 | |
| WebWeaverSystem category=Prompt-Based Frameworks, Base model=Qwen3-30B-A3B2026.02 | 0.2865 | 4.57 | 0.349 | 0.3646 | |
| Claude-researchSystem category=Proprietary Deep Research Systems2026.02 | 0.25 | 4.6 | 0.3889 | 0.3611 | |
| WebShaper-32BSystem category=Trained Open Models2026.02 | 0.0325 | 1.63 | 0.0375 | 0.93 | |
| OpenAI-deepresearchSystem category=Proprietary Deep Research Systems2026.02 | 0 | 5 | 1 | 0 | |
| openai-deepresearch2026.01 | 0 | 5 | 100 | 0 |