AgenticDataBench

What is AgenticDataBench?

Evaluates LLM-based data agents on realistic data science workflows across 15 domains, with fine-grained ground-truth labels and skill-level scoring.

Released

2026-07-02

Evaluates

General AI, Language & Knowledge, cs.DB

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2607.01647

AgenticDataBench paper and code

• Paper https://arxiv.org/abs/2607.01647
• Code https://github.com/AgenticDataBench/AgenticDataBench

Which sources cite AgenticDataBench?

• AgenticDataBench

Related General AI, Language & Knowledge, cs.DB benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench