Due to the continuously growing demand on pollutant exhaust gas reduction and the steadily growing pressure to improve engine fuel consumption, a further refinement of engine and electronic engine (combustion) controllers can certainly be expected in the future. In parallel the enabling technologies such as improved micromechanics, sensor technology and higher processing power at reasonable cost, will lead to new demands on the HIL technology. These necessities have an important impact on the models required for developing and testing modern engine control systems. The paper introduces the in-cylinder pressure-based engine simulation models as the next step for HIL simulation. It describes the parameterization process of this new type of engine model, which cannot be done in a straightforward manner. An optimization process is necessary to obtain correct model parameter since the physical process and the respective simulation model are highly transient, even at constant engine operating points. Special emphasis is placed on the software structure which separates the optimization process itself from the model-specific parts. This modular setup of the parameterization process allows fast implementation and testing of new modeling approaches like different heat release functions or wall heat flux models. With the current version of the parameterization process, gasoline and Diesel engine simulation models can be parameterized with the same optimization process. The paper describes the different optimization tasks for calculating model parameters with simplified submodels. Comparison between simulation results and in-cylinder pressure measurement curves show good correlations. Thus, an automatic model calibration is available for all types of in-cylinder pressurebased engine models. On the one hand, this allows accurate model parameterization even for this kind of transient models, and on the other hand it is possible to stay within the limited time schedule for model calibration in today's ECU test system projects. Thus, a fast parameterization technique which results in good closed-loop behavior in a reasonable project time has been developed. In the authors' view, in-cylinder models will replace the MVEM for HIL testing in the future. Additional sensor information such as in-cylinder pressure has to be provided for ECU test systems to satisfy state-of-the-art engine control units. Moreover, new actuators (variable valve train, multiple injection pulses, variable turbine geometry, etc.) increase the number of physical degrees of freedom in combustion engines. Physically based models promise the ability to simulate a reasonable system behavior without numerous test bench runs. With the increasing number of physical degrees of freedom and the inclusion of the transient system behavior, an automated process for model parameterization as presented in this paper is essential. The approach also allows the simulation model to be used as a kind of virtual test bench. It should also be possible to apply this general optimization approach to the parameters of a software ECU or a real ECU to obtain not only model but also optimum control algorithm parameters. This should make it possible to perform basic investigation of the engine behavior or to obtain starting parameters for an ECU calibration, which will reduce the time and cost of fine-tuning ECU parameters on a real engine test bench.