We present a dataset, DanFEVER, intended for multilingual misinformation research. The dataset is in Danish and has the same format as the well-known English FEVER dataset. It can be used for testing methods in multilingual settings, as well as for creating models in production for the Danish language.