Heat pumps are a vital part of each building for their role in keeping the space conditioned for the occupant. This study focuses on developing a model for the ground-source heat pump at the Zero Energy lab at the University of North Texas, and finding the minimum data required for generating the model. The literature includes many models with different approaches to determine the performance of the heat pump. Each method has its pros and cons. In this research the equation-fit method was used to generate a model based on the data collected from the field. Two experiments were conducted for the cooling mode: the first one at the beginning of the season and the second one at the peak of the season to cover all the operation conditions. The same procedure was followed for the heating mode. The models generated based on the collected data were validated against the experiment data. The error of the models was within ±10%. The study showed that the error could be reduced by 20% to 42% when using the field data to generate the model instead of the manufacturer’s catalog data. Also it was found that the minimum period to generate the cooling mode model was two days and two hours from each experiment, while for the heating mode it was four days and two hours from each experiment.