国产数据库跑出AI新能力！OceanBase登顶国际Data Agent榜单
Ranking
Overall
64
Content
70
Popularity
N/A
No observed public metrics; popularity remains neutral/archived.
Merged summary
TL;DR - OceanBase’s Scout Data Agent achieved 90.62% accuracy on the Data Agent Benchmark, ranking first and becoming the first submitted system to exceed 90%. The result highlights how combining an LLM, agentic workflows, and database infrastructure can support complex, real-world data analysis.
- Scout uses OceanBase with the GLM-5.2 model and reportedly outperformed entries built on GPT and Claude models.
- Unlike Text-to-SQL tests, DAB evaluates end-to-end data discovery, analysis planning, querying, computation, and answer verification across multiple domains and database types.
- The system builds data profiles with DataLens, plans execution paths, traces evidence, validates answers, and revises its approach when checks fail.
- Scout’s capabilities are expected to be incorporated into OceanBase DataPilot, the company’s AI-oriented data analysis product.
Sources (1)
国产数据库跑出AI新能力！OceanBase登顶国际Data Agent榜单
Public signals
N/A
TL;DR - OceanBase’s Scout Data Agent achieved 90.62% accuracy on the Data Agent Benchmark, ranking first and becoming the first submitted system to exceed 90%. The result highlights how combining an LLM, agentic workflows, and database infrastructure can support complex, real-world data analysis.
- Scout uses OceanBase with the GLM-5.2 model and reportedly outperformed entries built on GPT and Claude models.
- Unlike Text-to-SQL tests, DAB evaluates end-to-end data discovery, analysis planning, querying, computation, and answer verification across multiple domains and database types.
- The system builds data profiles with DataLens, plans execution paths, traces evidence, validates answers, and revises its approach when checks fail.
- Scout’s capabilities are expected to be incorporated into OceanBase DataPilot, the company’s AI-oriented data analysis product.