The Sumani River is an important water resource used for agriculture and domestic purposes. The river is also the main water supply for Lake Singkarak and used for a power plant. Proper water resource management is required for sustainable water availability for all water users. A hydrological model is a necessary tool to assess water management in the watershed. However, most models require several data sources not readily available in developing countries that describe the internal structure of the watershed. Artificial neural networks (ANN) are biologically inspired computer programs designed to simulate the way in which the human brain processes information and are capable of modeling a nonlinear system. River flow is an indicator of water availability in the watershed. It is greatly influenced by rain, so it has a pattern for its intensity. The objective of this study was to develop an artificial network model to predict river discharge in the Sumani River from rainfall and discharge patterns. Multiple Layer Perceptron with the backpropagation algorithm was applied to predict river flow. Data on rainfall and discharge from the preceding day from 2008 to 2012 were used to train the model, and data from 2013 to 2014 were used to test the model. The Sumani River ANN scheme can be used to predict river flow with correlation coefficients of 0.95 and 0.92 in the training and testing stages.