ThereareanumberofRainfallRunoffmodelsthatusedtopredictthefuturerunoffvolume of a given catchment. But the result of each models above and under the observed flow, those evaluation of hydrologic model behavior and performance is commonly made and reported through comparisons of simulated and observed variables. mostly, comparisons are made between simulated and observed stream flow at the catchment outlet. In this study, a daily rainfall- runoff Model modeling is very helpful to evaluate the performance of each models and analysis the sensitive parameter of them for GilgelAbaycatchmentbyusingfivelumpedconceptualmodelsthatareAWBM,Sacramento,SIMHYD,SMARandTANKarechecktheirperformancewithRainfall-Runoff Library. The calibration and validation of the RRL models were performed using SCEUA and Pattern search optimization methods together with Nash Sutcliffe criteria and runoffdifference%asprimaryandsecondaryobjectivesrespectively. Thequalityoffit betweentheobservedandsimulatedflowisjudgedbystasticalvaluesandvisualobservation of scatter plots of observed vs simulated flow correlation and hydro-graph. Furthermore, performance of the models were assessed by using Nash Sutcliffe efficiency criteria.all of the models are applied to the catchment, using split record evaluation, involving the calibration and validation periods (about 70% for calibration and 30% for verification). model sensitivity analysis was undertaken to analyze the sensitivity of model parameters of the selected model with regard to its objective function and subsequently the most sensitive parameters for the model were determined. Based on the resultsallmodelscanrepresenttheobservedflowinaverygoodperformance. whereas, in a validation period AWBM and SIMHYD models, have a very good performance to represent the peak and low flow and over predict intermediate flow of the catchment because those models work on saturation and infiltration excess respectively. where as, Sacramentomodelhasalimitationtocapturelowflow.soexceptSacramentomodelthe restfourmodelscanbeusedtopredictaGilgelAbayriverflowconditionsforhydraulic design and other research’s that uses flow data. Generally, AWBM model predicts the stream flowfairlywellusing SCE-UAoptimizationmethodwith overallNashSutcliffe Efficiency of 0.86 and 0.81 for calibration and verification periods respectively.Base flow index BFI (base flow index) and KSurf (daily surface flow recession constant) are the most sensitivity parameters of AWBM. Generally, conceptual models have given encouraging results in Gilgel Abay catchment.