This collection contains various language models that map arbitrary subsequences of text into embeddings matching Wikidata embeddings.