Unique challenges remain in the development of new drugs for the treatment of TB. While existing multidrug treatment regimens are prolonged and difficult for patients to adhere to, they are highly efficacious, setting a high bar for the performance of new agents. Complicating matters more, regulatory standards have changed since the first drugs for TB were introduced, with a rigorous characterization of the effect of a new drug within a combination regimen expected. If these demands are to be satisfied, innovative models will be needed to demonstrate drug efficacy. In the past, mycobacterial cultures performed on solid media at the end of treatment have been used as critical biomarkers of drug efficacy, but their inability to predict long-term outcomes with precision has limited their utility. This article reviews a range of nonclinical and clinical models to characterize the bactericidal and/or sterilizing activity of new compounds. Novel approaches, using in vitro and animal models, sensitive biomarkers, as well as creative new clinical trial designs, are discussed. These promise a timely expansion of our TB treatment armamentarium to include potent new drugs and shorter, simpler treatment regimens.