The energy consumption of Ventilation and Air Conditioning (VAC) systems in subway stations has attracted much attention, aiming to figure out energy-saving potentials and improve energy efficiency. Benchmarking is a useful method to evaluate energy performance of VAC systems during the operational stage. This study describes the development of an energy benchmarking model for the analysis of VAC energy consumption in subway stations with platform screen door systems in a variety of climate regions in China. Firstly, the Constraint Value (CV) and Leading Value (LV) of VAC energy use were simulated using a validated theoretical energy model, which represent the “general” and “superior” level of the VAC energy use, respectively. Results show that, for the 2,520 simulated cases, CV ranges from 184 to 859 × 103 kWh/year and LV ranges from 178 to 529 × 103 kWh/year, with different ranges for cold, hot-summer cold-winter, and hot-summer warm-winter region. Next, the simulated values of CV and LV were applied to establish an energy benchmarking model using the ordinary least square approach. For each climate region, the regressive energy benchmarking models for CV and LV were established and validated by cross-validation. Results show that R2 are 0.753–0.903 and MBE of the test set are 2.1–7.1%, indicating a good fit for the regressive benchmarking models. In order to further test their application, the regressive models were then applied to existing subway stations. Finally, the impacts of service-related variables, i.e. number of passengers, train departure density, and tunnel length, on the CV and LV were discussed. The proposed energy benchmarking model and analysis provide a convenient and effective tool for operators and managers to evaluate VAC energy consumption status and assess energy-saving potentials for subway stations.