Abstract Extreme flood events have regional differences in their generating mechanisms due to the complex interaction of different climate and catchment processes. This study aims to examine the capability of climate drivers to capture year-to-year variability in global flood extremes. Here, we use a statistical attribution approach to model seasonal and annual maximum daily discharge for 7,886 stations worldwide, using season- and basin-averaged precipitation and temperature as predictors. The results show robust performance of our seasonal climate-informed models in describing the inter-annual variability in seasonal and annual maximum discharges regardless of the geographical region, climate type, basin size, degree of regulation, and impervious area. The developed models enable the assessment of the sensitivity of flood discharge to precipitation and temperature changes, indicating their potential to reliably project changes in the magnitude of flood extremes.