In mouse models of tuberculosis infection, oral rifapentine-based regimens have demonstrated superior efficacy compared to current treatment options. However, these results were not replicated in recent human clinical trials due to limited rifapentine access into pulmonary granulomas. A novel inhalable rifapentine dry powder formulation could improve pulmonary rifapentine concentrations to rapidly treat infection. The pharmacokinetics of rifapentine in healthy mice was compared after intratracheal (IT) and intraperitoneal (IP) administration. Female BALB/c mice received rifapentine by IT or IP administration (20 mg/kg). Plasma, bronchoalveolar lavage (BAL) and tissue samples were harvested at each pre-specified time-point up to 24 hours and quantified. Concentration-time data were analysed using a mixed effects modelling approach to provide model-based estimates of area under the curve from time 0 to infinity (AUC0-∞). Whilst rifapentine was not detected in the BAL of mice dosed IP, IT delivery resulted in a maximum BAL fluid rifapentine concentration (25.2 ± 6.4 µg/mL) and an AUC0-∞ of 204.1 ± 67.8 mg/L.h. Between IT and IP delivery, the former had a considerably higher peak rifapentine lung concentration (321.3 ± 99.3 and 3.3 ± 1.2 µg/g, respectively) and AUC0-∞ (2614.4 ± 928.1 and 72.7 ± 24.9 mg/kg.h, respectively). The plasma AUC0-∞ after IT dry powder delivery (455.1 ± 132.4 mg/L.h) was approximately 4-fold smaller than the IP value (2010.1 ± 589.0 mg/L.h). Inhaled delivery of rifapentine has the potential to selectively enhance therapeutic efficacy at the pulmonary site of infection whilst minimising systemic exposure and related toxicity.