FLORES English-to-German

BLEU score for English-to-German translation on the FLORES-200 devtest set, as reported in this repository's model cards.

Also known as: flores200 eng_Latn-deu_Latn

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorytranslation
Subcategoryen-de
Page statusactive
MetricBLEU
Directionhigher_is_better
Unit%
Dataset size1012
Dataset licenceCC BY-SA 4.0
PublisherMeta AI (FAIR), NLLB Team; now maintained by the Open Language Data Initiative (OLDI)

What it measures

flores_en_de scores how well a model translates the FLORES-200 devtest sentences from English into German, an established high-resource pair also covered by WMT and other standard MT benchmarks.

Task format

Translate the 1012 FLORES-200 devtest sentences from English into German; score against the human German reference.

Models reporting this benchmark

These figures come from the model cards, which carry one collection date per card and no per-score attribution. They are shown as reported, not as verified evidence.
ModelProviderScoreCard as of
GPT-4OpenAI67.32026-04
GPT-4 TurboOpenAI67.32026-04
GPT-4oOpenAI67.32026-04
GPT-4o (2024-05-13)OpenAI67.32026-04
GPT-4o (2024-08-06)OpenAI67.32026-04
GPT-4o (2024-11-20)OpenAI67.32026-04
GPT-4o miniOpenAI67.32026-04
Claude Sonnet 3.5Anthropic66.92026-04
Claude Sonnet 3.5 v2Anthropic66.92026-04
Gemini 1.5 FlashGoogle DeepMind63.22026-04
Gemini 1.5 Flash-8BGoogle DeepMind63.22026-04
Gemini 1.5 ProGoogle DeepMind63.22026-04
Gemini 2.0 FlashGoogle DeepMind63.22026-04
Gemini 2.5 FlashGoogle DeepMind63.22026-04
Gemini 2.5 ProGoogle DeepMind63.22026-04

Data

This page as JSON · Edit on GitHub