List your company

Deep Search and Research Reasoning on xbench DeepSearch 2510 (Pass@1)

75Pass@1 AccuracyTongyi DeepResearch
4955.7562.569.25Oct 28, 2025

Evaluation results

MethodLinks
Tongyi DeepResearchAgent Type=DeepResearch Agent, Configuration=30B-A3B2025.1075
DeepSeek-V3.1Agent Type=LLM-based ReAct Agent2025.1071
GLM 4.5Agent Type=LLM-based ReAct Agent2025.1070
Kimi ResearcherAgent Type=DeepResearch Agent2025.1069
OpenAI o3Agent Type=LLM-based ReAct Agent2025.1067
Claude-4-SonnetAgent Type=LLM-based ReAct Agent2025.1065
Kimi K2Agent Type=LLM-based ReAct Agent2025.1050