 The Gavian weirs serve the same functions as impervious weirs, but they have the advantage of being eco-friendly, more stable, and economical and low-to-medium headcases. Additionally, dissolved oxygen is one of the main factors used to determine the purity of water. To compare different models for predicting Gavian wear-aration performance efficiency, APE-20, this study has been conducted using multiple linear regression, MLR, neural network, NIN, neurofuzzy system, NFS, deep neural network, DNN, and reported empirical models. The results showed that the DNN model had the highest value of R2, 0.935, and NSE, 0.934, and had the lowest error rate in validation phase. In addition, the trapezoidal-shaped NFS model had the second best result with R2, 0.873, and NSE, 0.852, while the MLR model had the third best result with R2, 0.905, and NSE, 0.897. Finally, the remaining models of NFS-based and empirical relations were unable to perform. This article was authored by Nan Tuuri, K. M. Luxmi, and Subod Renjun. We are article.tv. Links in the description below.