Modeling is a common field in information technology. Since it is desired that these models work properly, a validation of them is necessary. The goal of this thesis is to establish a method to qualify model based applications. The validation process is done by testing the behavior of the application for certain models and analyzing the output. Given the large input state space of models, that cannot be exhaustively covered, we answer the question of how to minimize the number of test cases without a loss of safety. We search and find a way to create test models which test the behavior to qualify the corresponding model instance. Additionally, we find methods to reduce the number of test models by identifying the optimal choice of models which covers many application scenarios. During this, we also investigate how it is possible to automatize the testing process. We describe ways to automatically generate test models and discuss their advantages and disadvantages. Finally, we apply these results to Validas’ Configurable Report Generator to demonstrate the process. For this process, a mostly automatized test pipeline is provided.