In this article, the complete moment convergence for weighted sums of non-identically distributed negatively superadditive dependent (NSD) random variables is investigated. Some sufficient conditions of the complete moment convergence and the complete convergence for the maximum of weighted sums of NSD random variables are established. As applications, some equivalent conditions of the complete moment convergence for a weighted version of NSD cases are presented. Our main results extend and improve the corresponding earlier ones in the literature.