This article describes the nature of the host response to Mycobacterium tuberculosis in the mouse and guinea pig models of infection. It describes the great wealth of information obtained from the mouse model, reflecting the general availability of immunological reagents, as well as genetic manipulations of the mouse strains themselves. This has led to a good understanding of the nature of the T-cell response to the infection, as well as an appreciation of the complexity of the response involving multiple cytokine- and chemokine-mediated systems. As described here and elsewhere, we have a growing understanding of how multiple CD4-positive T-cell subsets are involved, including regulatory T cells, TH17 cells, as well as the subsequent emergence of effector and central memory T-cell subsets. While, in contrast, our understanding of the host response in the guinea pig model is less advanced, considerable strides have been made in the past decade in terms of defining the basis of the immune response, as well as a better understanding of the immunopathologic process. This model has long been the gold standard for vaccine testing, and more recently is being revisited as a model for testing new drug regimens (bedaquiline being the latest example).