Magnetic shape memory alloys (MSMAs) can display up to 10% recoverable strain in response to the application of a magnetic field or compressive mechanical stress. The amount of recoverable strain depends on the amount and direction of the applied stress and magnetic field as well as manufacturing, chemical composition, and training of the material. Due to their relatively large strains and fast response, MSMAs are suitable for a wide range of applications, including power harvesting, sensing, and actuation. The response of MSMAs is primarily driven by the reorientation of martensite variants. Power harvesting is possible due to this reorientation process and the accompanying change in material’s magnetization, which can be changed into an electric potential/voltage using a pick-up coil placed around (or on the side of) the specimen. The magnitude of the output voltage depends on the number of turns of the pick-up coil, the amplitude of the reorientation strain, the magnitude and direction of the biased magnetic field, and the frequency at which the reorientation occurs. This paper focuses on the ability of a two dimensional constitutive model, developed by the group to capture the magnetomechanical response of MSMAs under general two dimensional loading conditions, to predict the power harvesting output of a Ni2MnGa specimen. Comparison between model predictions of voltage output and experimental measurements of the same indicate that, while the model is able to replicate the stress-strain response of the material during power harvesting, it is unable to accurately predict the magnitude of the experimentally measured voltage output. This indicates that additional features still need to be included in the model to better capture the change in magnetization that occurs during variant reorientation.