The flood can cause wide destroy to property and life because of the supreme corrosive force and can be highly damaging. In order to decrease the damages cause by the flood, an Artificial Neural Network (ANN) model has been established to predict flood in Sungai Isap, Kuantan, Pahang, Malaysia. This model is able to imitate same as the brain thinking process and avoid any influence to the predict judgment. This study proposed Levenberg-Marquardt (LM) back-propagation with two different ratios that is (80%: 10%: 10%) and (70%: 15%: 15%) for training sample, testing sample, and validation sample. The data collected in terms of temperature, precipitation, dew point, humidity, sea level pressure, visibility, wind and river level data were collected from January 2013 until May 2015. The results are shown on the basic of mean square error (MSE) and regression (R). The prediction by Levenberg-Marquardt with 80% training sample was shown better result compared with 70% training sample.