Wiktionary:Thesaurus or Wikisaurus

This page only applies to the English-language Wiktionary. As far as we know there is no similar extraction for any of the other Wiktionaries. (We're also not sure if the other Wiktionaries even have thesaurus pages.)

English-language thesaurus entries in the English-language Wiktionary are extracted. This is mentioned by Sérasset in this news post from 2017. In the post he states these entries are extracted as pages (i.e., dbnary:Page) rather than lexical entries (ontolex:LexicalEntry). See Extraction-of-pages-(vocables) for details and links to examples.

Note that the project (in Wiktionary) is now called 'Wiktionary:Thesaurus'. It was previously called 'Wikisaurus' and the news post linked above and extractor code refers to it by its old name.

There are currently (September 2026) n=2925 English-language thesaurus entries. The languages with the next highest numbers are Chinese with n=2214 and Hindi with n=306 entries. We emphasize these latter counts still refer to the English-language Wiktionary. Current counts can be obtained here.

'See' and 'See also' references not extracted

Note that although thesaurus pages are extracted, references to the thesaurus from the main namespace pages are not. For example, consider https://en.wiktionary.org/wiki/amazing. The 'Synonyms' section of the adjective entry says: 'See Thesaurus:awesome'. However, this reference is not extracted into DBnary. The same is true in cases where the synonyms are specified at the sense level. For example, consider https://en.wiktionary.org/wiki/hot. The first subsense of the first sense for the adjective form gives the synonym 'heated' and then states 'See also: Thesaurus: hot'. But only 'heated' is extracted as a dbnary:synonym attribute for this sense.