Abstract: A general network model with growth and preference features is proposed according to basic mechanisms for evolution in complex networks. The model was analyzed via continuum dynamic theory. Networks generated according to a GP model followed power law distribution based on maximum likelihood theory, explaining the cause of formation of the power law phenomenon. In addition, the exponential cut-off degree distribution in networks and the evolution of Internet topology are researched using a GP model, analyzing the long term evolution of the Internet and the behavioral feature of nodes.