We distribute two models for language identification, which can recognize 176 languages (see the list of ISO codes below). These models were trained on data from Wikipedia, Tatoeba and SETimes, used under CC-BY-SA.