The model for lemmatisation of standard Croatian was built with the CLASSLA-StanfordNLP tool (https://github.com/clarinsi/classla-stanfordnlp) by training on the hr500k training corpus (http://hdl.handle.net/11356/1183) and using the hrLex inflectional lexicon (http://hdl.handle.net/11356/1232). The estimated F1 of the lemma annotations is ~97.6.
The difference to the previous version is that now it relies solely on XPOS annotations, and not on a combination of UPOS, FEATS (lexicon lookup) and XPOS (lemma prediction) annotations.