Abstract: Exposure to per- and polyfluoroalkyl substances (PFAS) in drinking water sources seriously threatens water ecological security and human health. Due to the diversity of PFAS compounds, comprehensive control is costly. Currently, there is a lack of research on the screening technologies for priority control of PFAS, which restricts the precise risk management and control of PFAS. This study developed a screening method for priority control of PFAS based on the ToxPi (toxicological priority index) model, comprehensively considering multi-dimensional information, such as health risk, ecological risk, toxicity data, detected concentrations and detection rates, etc., to determine the priority ranking of PFAS and the priority control list. Then we applied the model in the screening of priority control of PFAS in the Liuxi River. The results showed that: (1) The PFAS concentration in the Liuxi River ranged from 1.79 ng/L to 24.73 ng/L, mainly composed of perfluorobutyric acid (PFBA), perfluorooctane sulfonate (PFOS), perfluorobutane sulfonate (PFBS), perfluorovaleric acid (PFPeA) and perfluorohexanoic acid (PFHxA). (2) The health and ecological risks of PFAS in the Liuxi River were acceptable. (3) Based on the ToxPi model, PFOS, perfluorononanoic acid (PFNA), perfluoroundecanoic acid (PFUnDA), and perfluorododecanoic acid (PFDoDA) were selected as priority control PFAS, which requires more attention and strengthened control. This study expanded and developed a ToxPi-based model for screening priority control PFAS, and provided a rational basis for the priority control list of PFAS from the perspective of protecting drinking water safety. This model is beneficial for the screening and precise risk control of priority control emerging contaminants in river basins and can be further applied for the identification and selection of priority control emerging contaminants in the key industries and risk sources.