Machine Translation

Measured on PanLex, Tatoeba, Machine Translate Foundation, Apertium, NLLB-200, FLORES-200, SIB-200, translatewiki.net.

Observatories

8 observatories measure Machine Translation.

6,241 languages

Crowdsourced lexical translation database aggregating thousands of dictionaries; counts here are meaning–expression entries per language, from the "meanings" export.

423 languages

Crowdsourced sentence and translation database; counts here are sentence, translation-link, and audio-recording totals per language, from Tatoeba's own exports.

Machine Translate Foundation

Open index →

632 languages

A community reference for machine translation, cataloguing which languages the major translation APIs and models claim to support.

234 languages

A free and open-source rule-based translation platform, whose published language pairs concentrate on related and less-resourced languages.

196 languages

Meta's No Language Left Behind translation models covering ~200 languages (FLORES-aligned).

FLORES-200

Open index →

194 languages

Evaluation benchmark for ~200 languages — the de facto baseline for whether a language is evaluable at all.

197 languages

Topic classification benchmark across ~200 languages; among the widest eval coverage available.

translatewiki.net

Open index →

317 languages

The translation platform behind MediaWiki and other free-software projects, reporting per-language completion across the message groups it hosts.