Non-replicating persistent Mycobacterium tuberculosis is difficult to kill since the organisms become undetectable using our conventional diagnostic methods and tolerant to anti-TB drugs. Resuscitation-promoting factors (RPFs) have been used to 'wake up' non-replicating persisters, making them easy to detect. Bedaquiline is a novel bactericidal and sterilizing anti-TB drug with the potential to eradicate RPF-dependent persistent M. tuberculosis. We present the first head-to-head comparison between the standard anti-TB regimen and a bedaquiline-modified regimen in eradicating RPF-dependent persistent M. tuberculosis, using the well-defined Cornell Model. M. tuberculosis-infected mice were treated for 14 weeks with either the standard regimen (rifampicin, isoniazid, pyrazinamide and ethambutol) or the same regimen where ethambutol was replaced by bedaquiline. The efficacy of both drug regimens was measured by cfu count elimination and eradication of persistent bacteria, which was evaluated using culture filtrate (CF) containing RPFs. At the end of treatment, the remaining cfu count-negative mice were administered hydrocortisone for 8 weeks. The induced disease relapse rates were determined by the percentage of mice that became positive for M. tuberculosis in the lung, spleen or both. The bedaquiline-containing regimen achieved total organ cfu count clearance at 8 weeks after treatment initiation, faster than the standard regimen (14 weeks). Importantly, the bedaquiline-containing regimen removed CF-dependent persistent bacilli at 8 weeks, leading to no disease relapse. A bedaquiline regimen eradicated persistent TB infections and completely prevented disease relapse in mice. These findings offer the potential for a faster cure for TB, with reduced relapse rate.