Prior to the HL-LHC operation, detector simulations undergo further developments in order to adapt to the increasing amounts of events. Current infrastructure still faces limitations to tackle the expected increase in data in terms of storage capacity and computation time. Early event reconstructions were based on matrix element methods followed by Monte Carlo techniques. Recent advances focused on boosting the speed of event simulations. For instance, C++ - based DELPHES uses simplified detector geometries and particle-material interactions, hence mapping the detector response into a parametric function. Nevertheless, this technique is limited by its exclusiveness to a specific detector geometry at a time, with any detector changes requiring the framework to be adjusted accordingly through hand-coding. Falcon (previously Turbosim), a fast stimulation framework that uses non-parametric methods to discern detector responses without the need for hand-coding. That being said, we aim to investigate the efficiency of deep generative models in simulating event reconstructions in a given detector, hence potentially replacing conventional complex algorithms.