The non-linear deformation analyses for problems involving liquefaction have been a major area of interest and research for the geotechnical earthquake community because there is currently no standard methodology that is widely acceptable for such problems, despite the increased sophistication the development of the theoretical background and computational procedures. This is particularly important because it denotes the fact that although the mechanics behind liquefaction have been studied extensively over the last years, there is still a great deal of uncertainty associated with its prediction and evaluation.;Herein, an evaluation of the effects of different constitutive models on the results of nonlinear deformation analyses and the effects of input parameter variations for specific protocols on model calibration and numerical procedures is attempted.;Two case histories of liquefaction will be modeled using the finite difference software FLAC, and their response to seismic loading will be evaluated using three constitutive models. The three constitutive models to be used are a pore-pressure generation model developed by URS, UBCSAND and PM4SAND developed by Boulanger (2010) following the basic framework of the stress-ratio controlled, critical state compatible, bounding surface plasticity model for sand initially presented by Manzari and Dafalias (1997) and later extended by Dafalias and Manzari (2004) and developed with its primary purpose being to address the need in geotechnical earthquake engineering practice for a constitutive model that captures the complex liquefaction related phenomena and at the same time can be easily calibrated to the engineering design relationships that are pertinent to the problem studied.;The predicted response of each model along with the calibration process will be documented and an attempt will be made to quantify some of the uncertainty in the response associated with the input parameters. The goal of the research is to provide a consistent framework for calibrating a model and some insight into how additional complexity in the model affects the predicted response.