Evaluating the relative importance of deterministic and stochastic processes underlying taxonomic and functional beta diversity is crucial in community ecology, because it can reveal the dominant processes of community assembly. However, studies of bird communities remain rare and of limited spatial extents. In this study, we described the taxonomic and functional beta diversity patterns of 32 passerine bird assemblages of Yunnan Province, China. We constructed null models based on observed species beta diversity and used multiple regressions on distance matrices to evaluate the relative contributions of deterministic and stochastic processes to passerine bird assemblage dissimilarity. Our results showed significant geographic distance decay in taxonomic and functional similarity, with passerine bird assemblages located in the northwest and southwest of the province having higher functional beta diversity values than expected. Environmental distance and geographic distance explained a similar amount taxonomic beta diversity, but environmental distance explained much more functional beta diversity. Our results suggest that both deterministic and stochastic processes drive taxonomic beta diversity, but that deterministic processes, particularly environmental filtering, play a dominant role in driving functional beta diversity of passerine bird assemblages at sub-national scale.