AWS is an Adaptive Weights Smoothing package. A reference implementation of the adaptive weights smoothing procedures (AWS) is available in form of a contributed package of the R-Project for Statistical Computing (http://www.r-project.org/ ). The package includes functions for local polynomial structural adaptive smoothing in regression models with additive errors and for local constant structural adaptive smoothing in exponential family models, the latter including binary response, Poisson regression, exponential regression and volatility models. The special case of a grid design allows for efficient reconstruction of non-smooth images in 2D and 3D. The package can be obtained from http://CRAN.R-project.org/ . An extension of this package currently allows for multivariate models and contains functions for the analysis of fMRI and dMRI data.