// BENCHMARK

xbench-DeepSearch benchmark

AI model leaderboard for the xbench-DeepSearch benchmark. Compare how large language models score on xbench-DeepSearch, see the full ranking, and understand what this AI benchmark measures. Tongyi DeepResearch currently leads with 75. Agentic deep-search benchmark from xbench — evaluates multi-source web research and synthesis.

Leaderboard

# Model Organization Score Variant Source
#1 Tongyi DeepResearch Alibaba-NLP 75 react-avg@3 official ↗
#2 DeepSeek-V3.1 DeepSeek 71 react-cited-tongyi official ↗
#3 GLM 4.5 Zhipu AI 70 react-cited-tongyi official ↗
#4 o3 OpenAI 67 react-cited-tongyi official ↗
#5 Claude Sonnet 4 Anthropic 65 react-cited-tongyi official ↗
#6 Kimi K2 Moonshot 50 react-cited-tongyi official ↗

Frequently asked questions about xbench-DeepSearch

What is the xbench-DeepSearch benchmark?

Agentic deep-search benchmark from xbench — evaluates multi-source web research and synthesis.

How is the xbench-DeepSearch benchmark scored?

xbench-DeepSearch is scored using the accuracy metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.

Which AI model scores highest on xbench-DeepSearch?

As of the latest reported scores on GenAIList, Tongyi DeepResearch achieves the highest result on xbench-DeepSearch with a score of 75.

Is a higher xbench-DeepSearch score better?

Yes. On xbench-DeepSearch a higher score indicates better performance, so models near the top of the leaderboard are the strongest.