Abstract: Wellbore liquid loading is one of the problems faced in the production process of gas wells. Liquid loading will cause the yield of gas wells to decrease, and even cause the off production of gas wells in severe cases. Accurately predicting the critical liquid-carrying gas flow rate of a gas well can help producers take timely measures to prevent the occurrence of liquid loading. Through comparing the minimum pressure gradient model, droplet model and liquid film model, and analyzing the results of the liquid loading experiment, it is indicated that the reverse movement of liquid film is the main reason for gas well liquid loading. According to the velocity profile of the liquid film in different ranges of gas velocity, this paper defines the gas superficial velocity corresponding to the zero shear stress between liquid film and pipe wall as the critical gas velocity of gas well liquid loading. Based on the annular mist flow pattern and taking into account the influence of such factors as pipe diameter, liquid flow rate and also droplet entrainment in the gas core, this study develops a zero shear stress model suitable for predicting the liquid loading of vertical gas wells. The new model and the existing prediction models of liquid loading are compared and verified by experimental data and field data; and the prediction accuracy and error of the model are used as evaluation indicators. The results show that the prediction agreement of the new model is better than other models, and the new model based on zero shear stress can predict the gas well liquid loading more accurately.