Identifying the main taxonomic, phylogenetic and trait dimensions of beta diversity, and evaluating their prospective drivers, advances our understanding of patterns and processes involved in the evolution of biological assemblages. Using comprehensive databases on the distribution, phylogeny, and morphological traits (later referred as functional traits) of Amazonian freshwater fishes, we analyzed beta diversity patterns of these three dimensions to evaluate prospective historical and contemporary drivers. We mostly focused on the pure turnover components of these three beta diversity dimensions (Taxoβsim, Phyloβsim, Traitβsim) and related them to Amazon Basin-wide predictors using multiple regression on distance matrices. We found mean taxonomic beta diversity about two times higher than mean phylogenetic and six times higher than species traits beta diversity, and coincident spatial patterns in Taxoβsim and Phyloβsim dimensions, whereas Traitβsim seemed more diffuse and heterogeneous across space. Our models revealed the prominent influence of sub-basins geographic distances, habitat harshness and water color types on the taxonomic and phylogenetic dimensions of beta diversity, together with smaller individual effects of current temperature and habitat types, historical sub-basins connections and marine incursions, and sampling effort. By contrast, Traitβsim was weakly explained only by sampling effort and current sub-basins hydro-morphological conditions. These results point to leading effects of dispersal limitation, environmental filtering and historical contingencies in explaining Amazonian fish assemblages taxonomic and phylogenetic beta diversity patterns, but not functional traits turnover.