Accurate estimation of side weirs’ discharge coefficient is an important issue in practical and economical design of these structures. In this study, the discharge capacity of labyrinth side weirs was estimated using adaptive neuro-fuzzy inference system (ANFIS) model. The target output of the present study was the discharge coefficient of labyrinth side weirs. Correlation tests indicated that the governing dimensionless parameters are almost independent. Results also showed that the apex angle of labyrinth weir is the most effective parameter on discharge coefficient compared with the others. The results of error analysis indicated that among different methods developed for estimating the labyrinth weir discharge coefficient, the relations presented by Emiroglu et al. (2010) and the developed ANFIS model in this study had the maximum and minimum error values, respectively. It is observed that the developed ANFIS model, presented more accurate results compared with the relations suggested by the other researchers for estimating the discharge coefficient of triangular labyrinth side weirs.