Heat capacity is a crucial parameter for process design. Ionic liquids (ILs) offer the opportunity to tailor the relevant properties for many processes, but heat capacity data often are not available. In this work, temperature-dependent contributions to the heat capacity for 39 cations and 32 anions are proposed. A database containing 104 ILs with 2443 data points was used for the development of this model. The experimental data in the test set could be reproduced with a mean relative error of 4.4%. This allows for a systematic screening of a huge IL database to determine the solvent best-suited for a task without the necessity to measure all the missing data. Furthermore, an optimization taking different substance properties into account by using respective mathematical tools is possible by combination with prediction models for other properties.