In this study, thermomechanically coupled finite element analysis (FEA) was performed for forming aluminum rectangular cups at elevated temperatures. In order to identify the onset of a failure during FEA, applicability, accuracy, and repeatability of three different failure criteria (maximum load, minimum thickness, and thickness ratio) were investigated. The thickness ratio criterion was selected since it resulted in accurate prediction of necking-type failure when compared with experimental measurements obtained under a variety of warm forming conditions. Predicted part depth values from FEA at various die-punch temperature combinations and blank holder pressures conditions were also compared with experiments, and showed good agreement. Forming limit diagrams were established at three different warm forming temperature levels (, , and ). An increasing limiting strain was observed with increasing forming temperature both in FEA and experiments. In addition, strain distributions on the formed part obtained under different die-punch temperature combinations were also compared to further validate the accuracy of FEA. A high temperature gradient between die and punch was found to result in increased formability; i.e., high part depths.