Aerosol prediction and aerosol–climate interactions have been incorporated into the China Meteorological Administration Climate Prediction System version 4 (CMA-CPSv4), which was recently developed based on the fully coupled global climate–aerosol model Beijing Climate Center Earth System Model version 1 (BCC-ESM1). In this study, 20 years of 30-day hindcasts initialized on the first day of each month from 2005 to 2024 are conducted to assess the predictability of five aerosol species—dust, sea salt, sulfate, organic carbon (OC), and black carbon (BC)—and aerosol optical depth (AOD) at 550 nm on weekly and monthly timescales. A 20-year aerosol replay simulation constrained by observed meteorology is used as the reference. At the global scale, CMA-CPSv4 maintains useful predictive skill for four aerosol species, excluding sea salt, during the first two forecast weeks and the first forecast month. During weeks 3–4, predictive skill for dust and sea salt persists primarily over the tropics and parts of North America, southern India, and southeastern China. Sulfate remains predictable mainly over North America, Africa, southeastern China, and the tropics, whereas OC and BC retain skill across extensive global land areas and the tropics. Compared with aerosol burdens, AOD is more difficult to predict, with useful predictive skill beyond the first forecast week largely confined to tropical regions. Case studies further demonstrate that CMA-CPSv4 can capture the large-scale transport of regional dust and BC aerosols at lead times of approximately 1–2 weeks. However, notable limitations remain, including the systematic underestimation of dust column burdens over major dust source regions after the first forecast week and limited predictive skill for AOD and sea salt.