System forecast is the precondition of scientific decision-making and management optimization. As to grey forecast model, it is not only a new kind of forecasting approach, but also one of the core topics of grey sys? tem theory. This paper deeply deconstructs the modeling mechanism of grey forecasting models, summarizes the procedures of the grey forecast modeling process and, depending on characteristics of the models, dividing the grey forecast models into linear structure and nonlinear structures. Considering models’forms are continu? ous or discrete and the modeling variables are single variable or multivariable, grey forecast models are further subdivided into continuous single variable grey forecast model, discrete single variable grey forecast model, con? tinuous multivariable grey forecast model and discrete multivariable grey forecast model. In the last part of the paper, theoretical and practical progress of grey forecast models are summarized so as to present the frontiers of grey forecast models and applications.