Underground water sources are one of the most important sources of water supply, so their modeling is important. In the meantime, studying and investigating the fluctuations of the underground water level in terms of management studies, creating engineering structures, agricultural uses and obtaining high quality underground water is of great importance. Most of the demand for drinking and agricultural water in Mashgin Shahr plain is also supplied through underground water. Four artificial intelligence models, including forward neural network, backward neural network, Sagno fuzzy logic and support vector machine were used to predict the groundwater level.