In this review we address the animal models currently in use to study HIV-1 latency and persistence. We outline the rationale behind each model, the major scientific findings made, and discuss the extent of their relevance to HIV-1.Several animal models have allowed the complex area of viral latency to be studied in vivo. These models are, by necessity, multifaceted and often labor intensive, leading to a slow but steady generation of information. Studies in animals have confirmed many observations in humans, i.e. that residual virus persists in the presence of antiretroviral agents and that this virus is competent to resume replication. They have also shown that viral reservoirs are distributed among several anatomical sites, including peripheral blood, spleen, lymph nodes and the central nervous system, and these reservoirs can be reactivated by immune activation. Murine (severe combined immunodeficiency-human) studies have supported the hypothesis that latent T-cell reservoirs are generated by activated cells that subsequently transition to a quiescent phenotype. In the course of these studies, several novel activators of latent virus with clinical potential have been identified.The latent reservoir is still the most significant barrier to eliminating HIV-1 infection. The development of several animal models of infection is helping us to unravel the factors that allow HIV-1 to persist in vivo and to test therapeutic strategies to deplete this reservoir.