Benchmark
DataAnalyticsBench
Test whether analytics agents answer business questions while catching ambiguity, statistical traps, and data defects.
Published leaderboard
Explore scores and evaluation settings on the publisher’s leaderboard.
Open leaderboardCoverage
43 English questions over one synthetic wind-energy warehouse, spanning analytics, ambiguity, impossible requests, statistics, data quality, and hidden patterns.
Methodology
Agents use Plotly Studio Embedded tools. A judge checks expected answers or required caveats with no partial credit. Plotly credits collaboration with researchers at Mila.