In this paper, exponential synchronization for CohenGrossberg neural networks with nonlinear impulsive effects is discussed. By Lyapunov functinoal method and by utilizing inequality techniques, some new and useful sufficient conditions on the exponential synchronization are obtained based on rnorm. Compared with recent years of linear impulsive effects results about neural networks synchronization, our results remove the restrictions of the impulsive gain γik∈ ［-2,0］, so our results are more general.