Modeling of water flowing into dam reservoir is one of the most important steps in watershed management, exploitation of dams, flood warning systems, and priority areas for erosion and sedimentation. In fact, optimal management of water resource systems, such as dams requires the accurate prediction of inflow into the dam reservoir. Therefore, it is essential to estimate this parameter more accurately. Several methods have been developed to predict the dam reservoir inflow. In the current study, the intelligent Support Vector Machine (SVM) and Nero-Fuzzy Adaptive Inference System (ANFIS) methods are used to estimate the inflow rate of the Sattarkhan dam and the effect of different input parameters such as monthly precipitation, discharge and temperature on improving the models accuracy is investigated. The results showed the desired efficiency of the Meta model approaches in estimating the monthly inflow into the Sattarkhan dam reservoir. The best results for the test data, in the state of modeling based on monthly discharge and precipitation was obtained the values of R= 0.878 DC= 0.782, RMSE= 0.063 and in the state of modeling based on monthly temperature, precipitation and discharge was obtained the values of R= 0.805, DC= 0.708 and RMSE= 0.108 were obtained. According to the results, the model with the parameters of the monthly discharge and precipitation leads to more accurate results.