This paper presents the development of an optimization strategy for ground source heat pump (GSHP) systems equipped with variable speed pumps in the ground loop system. The optimization problem is formulated using a model-based approach, in which the component models are used to estimate the system performance under various trial settings and an exhaustive search method is used to identify the optimal settings under the search ranges defined. The variable optimized is the outlet water temperature from the ground heat exchangers, which can be used as a set-point to control the operation of the variable speed pumps in the ground loop system. The overall objective of the optimization is to minimize the system power consumption while providing required building heating and cooling demand. The performance of the proposed strategy is tested and evaluated through simulations. It is shown that, compared to a two-stage control strategy for variable speed pumps, the proposed strategy can save 4.2% of cooling power consumption of the GSHP system studied. The methodology used in the development of this proposed strategy can be potentially useful for control optimization of any types of GSHP systems. For complex systems, a performance map can be generated based on this method and then used to practically control the operation of GSHP systems.