The transparency and open availability of energy system models and their input data are of particular importance due to their increasing complexity and policy relevance. In recent years, a large number of model-based scenario analyses have been carried out. These analyses are based on diverse model approaches and lead to a rather broad range of results, which due to different data structures and mathematical approaches are hardly directly comparable. In this paper, detailed and harmonized scenario input parameters are the basis of a systematic model experiment including four electricity system models. In the following, the different model approaches are classified and their respective results are discussed transparently. Consequently, differences in results can be interlinked directly with model properties. The results are compared focusing on a selection of output parameters, such as investment and dispatch decisions in flexible power plants, storage dispatch, wholesale electricity prices, CO ₂ emissions and generation adequacy in hours with critical supply situations in Germany until 2030. Differences in the results are traced back to conceptual differences as the models can be distinguished not only with regard to their mathematical approaches, but also to their level of detail. Results indicate that next to the differences of the mathematical approaches (i.e., linear optimization vs. agent-based simulation), the myopic foresight perspective (e.g., rolling planning algorithm with 24- and 36-hours loops vs. perfect foresight in a closed loop for one year) are decisive for the range of obtained results.