The development of reciprocating compressors and their systems requires models that accurately predict the performance at various operating conditions. Numerous mathematical models for compressors can be found in the current literature, but many lack the flexibility to adapt to different refrigerants. This paper presents the further development of a semi-empirical compressor model, which is continuously adapted using measurement data and then examined for transferability to other refrigerants. The experimental data was obtained from a reciprocating compressors operated with different refrigerants. The model was developed in Modelica using the TIL library. Compared to purely empirical approaches, the semi-empirical model enables a more detailed mapping of relevant loss mechanisms, which allows performance potentials and limits to be identified at an early stage, development processes to be optimized and testing efforts to be reduced. The results presented are preliminary and show both the strengths and the current limitations of the model. Further investigations are required to validate the model’s transferability to a wider range of refrigerants