Efficient irrigation management is crucial to achieve optimal rice yields, and crop modeling serves as a valuable tool for optimizing irrigation schedules. In this respect, a field experiment was carried out at Bangladesh Agricultural University, Mymensingh, from January to June 2023 and 2024, to evaluate the AquaCrop model's accuracy in simulating yield and irrigation for dry direct-seeded rice under various irrigation managements. Three irrigation managements: no irrigation (I1), always remain field capacity (FC) moisture (I2) and irrigation at 25% of water disappearance from FC (I3) were applied to four rice varieties: BRRI dhan28, BRRI dhan29, BRRI dhan89, and BRRI dhan100. The model’s performance was assessed using the coefficient of determination (R²), root mean squared error (RMSE) and mean absolute error (MAE); with calibration achieving R² values of 0.98–0.99 for simulating canopy cover (CC) percentages and model validation achieving R² values of 0.96–0.98. Experimental and simulation results highlighted BRRI dhan89 as the most productive variety under the I2 treatment, achieving better yields (5.33 t ha⁻¹), biomass (12.83 t ha⁻¹), and water productivity (11.24 kg ha⁻¹ mm⁻¹) than other treatments. This model successfully simulated rice growth, yield, and irrigation requirements, demonstrating its reliability as a tool for future research on irrigation management and crop productivity in Bangladesh. This study supports AquaCrop's application for sustainable agricultural practices and water management in rice cultivation.