Active magnetic regenerator refrigeration is an alternative refrigeration technology to conventional vapor-compression refrigeration. This technology has potential for higher efficiency and less environmental impact in comparison to conventional vapor-compression refrigeration. In recent years substantial amount of studies have been done on modeling of active magnetic regenerators. Some of these models are time dependent models and consequently computation time is relatively high and therefore not suitable for online applications, advanced optimization and control of such a system. In this paper, a computationally efficient model of an active magnetic regenerator is derived and validated. The derived model is based on a data-based modeling technique not previously implemented for active magnetic regenerators. The derived model is numerically more efficient and computation time for system analysis, simulation, optimization and control reduces significantly by using this model. The results show that this model can capture the local nonlinear behavior of the system, and it is validated by comparing predicted performance with the full dynamic model.