This papera ddresses the rainfall forecasting in Manaus with different Machine Learning techniques. Considering this goal, we used monthly rainfall historical data from 65 years as well as Niño indexes as input into four forecasting approaches: decision trees, random forests, artificial neural networks and nearest neighbors. More than 3800 models were proposed, trained and tested from which it was possible to see that the best results were provided by neural networks, with an F-score of 70%. The results obtained can be used by several institutions for strategic decision making in order to minimize the negative effects of precipitation in this city.