In fluid dynamics, incipient motion generally refers to the point at which sediment particles, such as sand or gravel, begin to move in the flowing water streams. In this regard, the critical shear stress is an important parameter that determines whether sediment particles will be eroded, transported, or deposited in the fluid flow. By considering the importance of incipient motion, machine learning algorithms such as the Extra Tree Regression (ETR), Decision Tree (DT), and Gradient boosting Regressor (GBR) are used in the present study for predicting the critical shear stress in an alluvial channel. Further, the prediction of critical shear stress is carried out while considering the input variables such as acceleration due to gravity (g), depth of flow (y), friction slope (Sf), specific gravity (G), size of sediment (d), kinematic viscosity (ν), and velocity (u) with the help of these machine learning models. Thus, the performance of models is evaluated on the basis of the coefficient of determination (R2) where it was observed that the Decision Tree (DT) model performed better on the basis higher R2 value of 0.92 in comparison to the reaming models. In addition to this, with the help of prediction of the critical shear stress, it may contribute significantly to researchers and professional’s community in order to estimate erosion rates, sediment deposition patterns, and potential impacts on infrastructure and ecosystems.