Crop simulation models are broadly applied for improving crop production management. To estimate the performance of CERES-Maize model on maize variety single cross 704 under three levels of irrigation (I1: 70 mm evaporation, I2: 100 mm evaporation and I3: 120 mm evaporation), an experiment was conducted based on a randomized complete block design with four replications at Research Field of Faculty of Agriculture and Natural Resources, Islamic Azad University, Karaj Branch, Karaj, Iran, during 2012–13. CERES-Maize model provided very satisfactory estimates of grain dry matter, total dry matter and grain yield. Wilmot's index of agreement was higher than 0.6 for grain dry matter and total dry matter. RMSE estimated for grain dry matter, total dry matter and grain yield were calculated as 827.318–931.73, 2027.63–3953.338 and 295–823 kg/ha, respectively. R2 for grain dry matter, total dry matter and grain yield was at the range of 0.937–0.982, 0.799–0.844 and 0.972. CERES-Maize model provided unsatisfactory estimates of LAI. Thus, with respect to satisfactory simulations of CERES-Barley model, this model can be used for the research purposes and management programmes in Karaj region after calibration and the accuracy test of model.