The degradation of atrazine (ATZ), sulfamethoxazole (SMX) and metoprolol (MET) in flow-through VUV/UV/H₂O₂ reactors was investigated with a focus on the effects of H₂O₂ dosage and reactor internal diameter (ID). Results showed that the micropollutants were degraded efficiently in the flow-through VUV/UV/H₂O₂ reactors following the pseudo first-order kinetics (R² > 0.92). However, the steady-state assumption (SSA) kinetic model being vital in batch reactors was found invalid in flow-through reactors where fluid mixing was less sufficient. With the increase of H₂O₂ dosage, the ATZ removal efficiency remained almost constant while the SMX and MET removal was enhanced to different extents, which could be explained by the different reactivities of the pollutants towards HO^•. A larger reactor ID resulted in lower degradation rate constants for all the three pollutants on account of the lower average fluence rate, but the change in energy efficiency was much more complicated. In reality, the electrical energy per order (E_EO) of the investigated VUV/UV/H₂O₂ treatments ranged between 0.14–0.20, 0.07–0.14 and 0.09–0.26 kWh/m³/order for ATZ, SMX and MET, respectively, with the lowest E_EO for each pollutant obtained under varied H₂O₂ dosages and reactor IDs. This study has demonstrated the efficiency of VUV/UV/H₂O₂ process for micropollutant removal and the inadequacy of the SSA model in flow-through reactors, and elaborated the influential mechanisms of H₂O₂ dosage and reactor ID on the reactor performances.