AbstractBingham and Herschel–Bulkley (H–B) constitutive models are widely utilized for describing the rheological behavior of cement grouts. The Bingham model is commonly adopted in grouting design owing to its simplicity, whereas the H–B model provides higher accuracy by incorporating a flow index that accounts for cement grout's shear thinning behavior. This technical note examines the effect of rheological model selection, the Bingham or H–B model, on the prediction of cement grout propagation in rock fractures. A set of rheological test data on typical cement grouts with the water/cement ratio between 0.8 and 1.2 is used to determine the parameters of both models. Numerical simulations of cement grout propagation in both homogeneous and rough-walled fractures using the two models are conducted for comparison. The results show that using the Bingham model consistently underestimates grout propagation length and flow rate compared to the results using the H–B model. In homogeneous fractures, the difference in maximum filling ratio reaches about 30% at the initial stage of grouting and decreases over time. In rough-walled fractures, the difference in maximum filling ratio stabilizes at 10–20%. Furthermore, the influence of the shear-rate fitting range on the model performance was analyzed. The Bingham model achieves its highest predictive accuracy within the fitting range of [25, 250] s−1, whereas the H–B model exhibits low sensitivity to fitting range variations. These findings provide theoretical and practical guidance for selecting suitable rheological models to improve the accuracy and reliability of rock grouting design.