Deep Research on DeepResearch Bench
58.03RACE OverallDuMate-DeepResearch
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| DuMate-DeepResearchNote=Average scores across multiple runs2026.06 | 58.03 | 59.48 | 61.48 | 53.87 | 54.34 | — | — | — | |
| ZTE Nebula DeepResearch2026.06 | 57.27 | 58.37 | 59.76 | 54.06 | 54.66 | — | — | — | |
| iFlow-Researcher2026.06 | 57.08 | 58.24 | 59.74 | 53.24 | 55.05 | — | — | — | |
| Zhipu Deep Research2026.06 | 57.06 | 58.15 | 60.14 | 53.47 | 53.88 | — | — | — | |
| Xiaoyi DeepResearch 6.02026.06 | 57 | 58.58 | 59.38 | 53.58 | 53.99 | — | — | — | |
| Cellcog-Max2026.06 | 56.67 | 57.4 | 60.01 | 53.25 | 53.21 | — | — | — | |
| 1688AILab-DeepResearch2026.06 | 56.53 | 57.32 | 59.27 | 53.51 | 53.36 | — | — | — | |
| Octen DeepResearch2026.06 | 56.31 | 56.89 | 59 | 53.39 | 53.83 | — | — | — | |
| Grep Deep Research2026.06 | 56.23 | 56.82 | 58.92 | 53.38 | 53.44 | — | — | — | |
| NVIDIA-AIQ2026.06 | 55.95 | 56.9 | 58.49 | 52.89 | 53.43 | — | — | — | |
| MS-Agent DeepResearch2026.06 | 55.31 | 56.76 | 56.79 | 53.1 | 52.28 | — | — | — | |
| Cellcog2026.06 | 55.31 | 55.41 | 58.21 | 52.5 | 53.12 | — | — | — | |
| CMCC-DeepInsight2026.06 | 55.24 | 55.66 | 58.7 | 52.53 | 50.94 | — | — | — | |
| TrajectoryKit2026.06 | 54.92 | 54.1 | 57.9 | 52.91 | 52.72 | — | — | — | |
| Onyx Deep Research2026.06 | 54.54 | 54.67 | 56.43 | 53.08 | 52.02 | — | — | — | |
| Bodhi Deep Research2026.06 | 54.22 | 54.23 | 56.09 | 52.86 | 51.81 | — | — | — | |
| AgentDisCoBackbone=Claude-Opus-4.62026.05 | 54.02 | 53.38 | 56.65 | 53.11 | 51.53 | 93.56 | 89.88 | — | |
| Deep Dog 12026.06 | 53.52 | 53.14 | 56.1 | 51.83 | 51.18 | — | — | — | |
| RecallRadar Intelligence2026.06 | 53.19 | 53.91 | 53.53 | 52.18 | 52.38 | — | — | — | |
| DualGraphBackbone LLM=GPT-5, Knowledge Graph Usage=with KG2026.02 | 53.08 | 53.78 | 54.3 | 52.48 | 49.81 | 57.55 | 79.65 | — | |
| Gemini-2.5-ProAgent Configuration=DR2026.02 | 52.54 | 52.65 | 53.2 | 51.82 | 51.83 | 72.11 | 53.38 | — | |
| LiAuto Mind DeepResearch 1.52026.06 | 52.54 | 51.54 | 55.3 | 50.45 | 51.26 | — | — | — | |
| Tavily Research2026.06 | 52.44 | 52.84 | 53.59 | 51.92 | 49.21 | — | — | — | |
| ThinkDepth.ai2026.06 | 52.43 | 52.02 | 53.88 | 52.04 | 50.12 | — | — | — | |
| AgentDisCoBackbone=Gemini-2.5-Pro, Harness=true2026.05 | 52.11 | 51.89 | 53.43 | 51.87 | 50.45 | 89.55 | 69.65 | — | |
| DualGraphBackbone LLM=GPT-4.1, Knowledge Graph Usage=with KG2026.02 | 51.5 | 52.07 | 52 | 51.41 | 49.3 | 56.23 | 76.33 | — | |
| AgentDisCoBackbone=Gemini-2.5-Pro2026.05 | 51.44 | 51.23 | 52.49 | 51.57 | 50.39 | 89.06 | 63.94 | — | |
| DualGraphBackbone LLM=GPT-5, Knowledge Graph Usage=without KG2026.02 | 51.17 | 51.69 | 51.79 | 51.16 | 48.75 | 61.15 | 66.06 | — | |
| Salesforce AIR2026.06 | 50.65 | 50 | 51.09 | 50.77 | 50.32 | — | — | — | |
| LangChain Open Deep ResearchComponents=GPT-5 + Gensee Search2026.06 | 50.6 | 50.06 | 50.76 | 51.31 | 49.72 | — | — | — | |
| Gemini-2.5-Pro-Deepresearch2026.05 | 49.71 | 49.51 | 49.45 | 50.12 | 50 | 78.3 | 165.34 | — | |
| Gemini 2.5 Pro DeepResearchReproduced=true2026.06 | 49.71 | 49.51 | 49.45 | 50.12 | 50 | — | — | — | |
| DualGraphBackbone LLM=GPT-4.1, Knowledge Graph Usage=without KG2026.02 | 49.18 | 50.13 | 48.41 | 50.06 | 47.35 | 59.96 | 62.96 | — | |
| OpenAI-DR2026.02 | 46.68 | 47.27 | 45.55 | 48 | 45.93 | 73.18 | 31.7 | — | |
| Doubao-DR2026.02 | 46.62 | 47.67 | 45.21 | 48.16 | 45.07 | 51.87 | 35.12 | — | |
| KimiAgent Configuration=Res.2026.02 | 46.5 | 47.06 | 45.67 | 47.53 | 45.47 | 20.8 | 2.3 | — | |
| Openai-Deepresearch2026.05 | 46.45 | 46.46 | 43.73 | 49.39 | 47.22 | 75.01 | 39.79 | — | |
| OpenAI DeepResearchReproduced=true2026.06 | 46.45 | 46.46 | 43.73 | 49.39 | 47.22 | — | — | — | |
| UESTC-MBSE-RAAA2026.06 | 46.13 | 43.77 | 48.34 | 47.21 | 43.78 | — | — | — | |
| DR-Tulu2026.06 | 45.49 | 44.08 | 44.65 | 49.56 | 42.3 | — | — | — | |
| Claude-Research2026.05 | 45 | 45.34 | 42.79 | 47.58 | 44.66 | — | — | — | |
| Kimi-Research2026.05 | 44.64 | 44.96 | 41.97 | 47.14 | 45.59 | — | — | — | |
| Doubao-Research2026.05 | 44.34 | 44.84 | 40.56 | 47.95 | 44.69 | 52.86 | 52.62 | — | |
| Langchain-Open-Deep-Research2026.05 | 43.44 | 42.97 | 39.17 | 48.09 | 45.22 | — | — | — | |
| GPT-4.12026.02 | 43.08 | 42.85 | 41.96 | 44.62 | 43.77 | 85.75 | 3.84 | — | |
| LangChain-DR2026.02 | 42.92 | 41.95 | 44.63 | 43.75 | 43.08 | 59.83 | 61.29 | — | |
| NVIDIA AIQAgent Configuration=RA2026.02 | 42.48 | 42.01 | 41.81 | 43.8 | 43.01 | 46.01 | 4.41 | — | |
| PerplexityAgent Configuration=Res.2026.02 | 42.44 | 42.36 | 40.85 | 44.41 | 43.11 | 76.03 | 21.69 | — | |
| Claude-3.7Agent Configuration=Search2026.02 | 41.56 | 41.42 | 39.99 | 43.8 | 42.02 | 87.32 | 24.51 | — | |
| GrokAgent Configuration=Deep Search2026.02 | 41.01 | 40.55 | 38.82 | 43.56 | 42.83 | 68.47 | 7.28 | — | |
| Qwen-Agent2026.02 | 40.95 | 39.19 | 43.65 | 42.88 | 41.27 | 52.23 | 8.39 | — | |
| Sonar-Reason-Pro2026.02 | 40.74 | 40 | 39.06 | 43.14 | 42.3 | 37.31 | 6.72 | — | |
| Tongyi-DR2026.02 | 40.62 | 38.63 | 43.92 | 42.58 | 41.03 | — | — | — | |
| Sonar-Reason2026.02 | 40.53 | 39.73 | 38.87 | 42.79 | 42.26 | 51.19 | 11.6 | — | |
| Sonar-Pro2026.02 | 40.31 | 39.76 | 38.35 | 42.68 | 42.02 | 76.64 | 13.06 | — | |
| Sonar2026.02 | 37.29 | 36.26 | 34.51 | 40.46 | 40.61 | 69.59 | 8.11 | — | |
| Claude-3.5Agent Configuration=Search2026.02 | 34.32 | 33.43 | 32.37 | 36.98 | 36.34 | 93.8 | 8.96 | — | |
| GPT-4.1-mini2026.02 | 34.21 | 33.36 | 30.52 | 38.18 | 37.64 | 82.63 | 4.08 | — | |
| ASearcher-Web-7BFamily=Open Search2026.05 | — | — | — | — | — | — | — | 10 | |
| Claude-Opus-4.52026.06 | — | — | — | — | — | — | — | 50.6 | |
| Claude-Sonnet-4.52026.06 | — | — | — | — | — | — | — | 47.2 | |
| DecomposeR-8BFamily=Ours, Training Stage=SFT only2026.05 | — | — | — | — | — | — | — | 36.3 | |
| DecomposeR-8BFamily=Ours, Training Stage=SFT + RL2026.05 | — | — | — | — | — | — | — | 41.8 | |
| Gemini Deep ResearchFamily=DR agent2026.05 | — | — | — | — | — | — | — | 48.9 | |
| Gemini-3-Flash High2026.06 | — | — | — | — | — | — | — | 46.1 | |
| Gemini-3-pro High2026.06 | — | — | — | — | — | — | — | 49.6 | |
| GPT-5 + SearchFamily=LLM+search2026.05 | — | — | — | — | — | — | — | 50.7 | |
| GPT-5-mini High2026.06 | — | — | — | — | — | — | — | 50.7 | |
| GPT-5.2 High2026.06 | — | — | — | — | — | — | — | 52.2 | |
| OpenAI Deep ResearchFamily=DR agent2026.05 | — | — | — | — | — | — | — | 46.9 | |
| Perplexity Deep ResearchFamily=DR agent2026.05 | — | — | — | — | — | — | — | 42.7 | |
| Qwen3-8B + SearchFamily=LLM+search2026.05 | — | — | — | — | — | — | — | 34.4 | |
| Search-R1-7BFamily=Open search2026.05 | — | — | — | — | — | — | — | 9.7 | |
| Seed2.0 Lite2026.06 | — | — | — | — | — | — | — | 54.4 | |
| Seed2.0 Pro2026.06 | — | — | — | — | — | — | — | 53.3 | |
| Tongyi DeepResearch-30B-A3BFamily=Open DR2026.05 | — | — | — | — | — | — | — | 40.6 | |
| WebExplorer-8BFamily=Open search2026.05 | — | — | — | — | — | — | — | 36.7 | |
| WebThinker QwQ-32BFamily=Open DR2026.05 | — | — | — | — | — | — | — | 37.9 | |
| WebThinker-32B-DPOFamily=Open DR2026.05 | — | — | — | — | — | — | — | 40.6 |