Immunoinformatics approaches are widely used in a variety of application from basic immunological to applied biomedical research. Complex data integration is inevitable in immunological research. Thus, applications are often complex pipelines including multiple tools, and data sources with intensive pre- and post-processing. Non-unified in- and output interfaces of immunoinformatics tools make the development of such applications tedious. Therefore, we developed FRED2, an open source immunoinformatics framework offering easy and unified access to methods for epitope prediction and other immunoinformatics applications. FRED2 is implemented in Python and designed to be extendable and flexible to allow rapid development of advanced applications.