Pearl millet (Pennisetum glaucum (L.) R.Br.) is widely grown in some of the driest regions of the world, mainly drier tropics. Although it is easy to cultivate under semi-arid and arid regions, it still responds very favourably to slight improvements in growing conditions such as supplementary irrigation. Because this crop is mostly cultivated under water-limited conditions, there is a need to develop strategies to promote efficient water use, and this can be achieved through field experiments and or crop modelling. The AquaCrop model requires a minimum number of crop parameters, with the aim of balancing simplicity, accuracy, robustness and user-friendliness. In this study, we calibrate and validate the AquaCrop model for an underutilised crop, pearl millet under irrigation and rainfed conditions. Experiments were carried out in lysimeters with two varieties of pearl millet (GCI 17, improved variety; Monyaloti, local variety) during the 2010–11 season. Field trials were conducted during the 2008–09 and 2009–10 seasons. The field plot was under a line-source sprinkler with four replications. Lysimeter datasets and field data (2008–09) were used for parameterisation and calibration of the model, and validation was done with the 2009–10 dataset. The model was able to simulate canopy cover, biomass production, cumulative evapotranspiration and grain yield, but not soil-water content, for the two varieties of pearl millet under irrigation and rainfed conditions. The performance of the model in simulating soil-water content is moderate for this crop and needs to be improved.