Abstract. Guaranteeing the essential properties of a system early in the design process is an important as well as challenging task. Modeling languages such as the UML allow for a formal description of structure and behavior by employing OCL class invariants and operation preand postconditions. This enables the verication of a system description prior to implementation. For this purpose, rst approaches have recently been put forward. In particular, solutions relying on the deductive power of constraint solvers are promising. Here, complementary approaches of how to formulate and transform respective UML and OCL verication tasks into corresponding solver tasks have been proposed. However, the resulting methods have not yet been compared to each other. In this contribution, we consider two verication approaches for UML and OCL behavioral models and compare their methods and the respective workows with each other. By this, a better understanding of the advantages and disadvantages of these verication methods is achieved.