This is the reference corpus of standard Croatian. In its 3.0 version, which is accessible via noSketch Engine, it has 216.8 million tokens. In terms of annotation, the corpus is tokenised, lemmatised and tagged for MSDs (morphosyntactic descriptions).