AgML · Crop yield forecasting
CY-Bench dashboard
Benchmarking AI for global crop yield forecasting
Interactive results for Benchmarking the State of AI for Crop Yield Forecasting on CY-Bench: a standardized comparison of statistical baselines, process-based crop models, feature-engineered machine learning, tabular foundation models, and deep sequence models across maize and wheat, evaluated on overall, spatial, temporal, and anomaly skill.
Open a country on the map for local results, or use the global insight pages below for cross-country summaries. Modeling code · Dataset · ESSD paper · AgML
Explore by country
Click a blue country for mid-season and end-of-season model results
Loading map…
Select a country
Click a highlighted country on the map.
Global insights
Cross-country summaries · models, horizons, crops, and sample size
If you use CY-Bench or results from this dashboard, please cite the dataset paper (Kallenberg et al., ESSD 2026) and, for the AI model assessment, Kallenberg et al. Benchmarking the State of AI for Crop Yield Forecasting: A Global Assessment Across Modeling Paradigms. KDD 2027.