Low cardiorespiratory fitness (CRF) is associated with functional disability, heart failure and mortality. Left ventricular (LV) end-diastolic volume (LVEDV) has been linked with CRF, but its utility as a diagnostic marker of low CRF has not been tested. This multi-center international cohort examined the relationship between LV size on echocardiography and CRF (peak oxygen uptake [peak VO2] from cardiopulmonary exercise testing) in individuals with LV ejection fraction ≥50%. Absolute and BSA-indexed LVEDV (LVEDVi) were tested as predictors of low CRF and functional disability (peak VO2 <1100ml/min or <18 ml/kg/min) and compared against candidate measures of cardiac structure and function. 2876 individuals (309 endurance athletes, 251 healthy non-athletes, 1969 individuals with unexplained dyspnea, 347 individuals with heart failure with preserved ejection fraction) were included. For the entire cohort, LVEDV had the strongest univariable association with peak VO2 (R2 =0.45, standardized [std]β 0.67, p<0.001) and remained the strongest independent predictor of peak VO2 after adjusting for age, sex and BMI (stdβ 0.30, p<0.001). LVEDV was better at identifying low CRF than most established echocardiographic measures (LVEDV AUC 0.72; LVEDVi AUC 0.71), but equivalent to the E/e' ratio. The probability of achieving a peak VO2 below the functional independence threshold was highest for smaller ventricular volumes, with LVEDV and LVEDVi of 88ml and 57ml/m2 providing the optimal cut-points, respectively. Small resting ventricular size is associated with a higher probability of low CRF and functional disability. LV size is the strongest independent echocardiographic predictor of CRF across the health-disease continuum.