Open-Ended Deep Research on DeepResearchGym
90.71ClarityGemini-2.5-Pro-DeepResearch
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Gemini-2.5-Pro-DeepResearch2026.02 | 90.71 | 99.9 | 93.37 | 99.69 | 95 | 97.45 | 96.02 | |
| DualGraphBackbone=GPT-4.1, Knowledge Graph usage=true2026.02 | 88.14 | 99.99 | 92.37 | 99.98 | 91.86 | 93.52 | 94.31 | |
| DualGraphBackbone=GPT-4.1, Knowledge Graph usage=false2026.02 | 87.11 | 99.95 | 92.16 | 99.95 | 89.66 | 93.24 | 93.68 | |
| Claude-Research2026.02 | 86.67 | 96.88 | 84.41 | 96.56 | 26.77 | 90.22 | 80.25 | |
| LangChain-DeepResearchBackbone=GPT-4.12026.02 | 85.2 | 93.3 | 90.5 | 96.9 | 85.9 | 85.9 | 89.62 | |
| OpenAI-DeepResearch2026.02 | 84.9 | 98.1 | 89.8 | 97.4 | 88.4 | 89 | 91.27 | |
| Tongyi-DeepResearchBackbone=GPT-4.12026.02 | 77.5 | 72.7 | 83.1 | 86.4 | 16.8 | 62.5 | 66.5 | |
| Doubao-Research2026.02 | 68.85 | 93.12 | 83.96 | 93.33 | 84.38 | 83.12 | 84.46 | |
| Qwen-AgentBackbone=GPT-4.12026.02 | 64.7 | 82.4 | 82.7 | 81 | 64.7 | 60.4 | 72.65 |