Benchmark RadarAI BENCHMARK PROFILE
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
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
Addresses the lack of comprehensive benchmarks for automating data science workflows, providing fine-grained skill-level insights to guide agent development and selection.
Motivation
Data science aims to derive actionable insights from heterogeneous raw data, unlocking the value of the massive amounts of data generated in modern society.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.