Bemba translation models bem

4 models into Bemba, 5 from Bemba. Best scored into Bemba: English to Bemba (chrF++ 31.9). Best scored from Bemba: Bemba to English (chrF++ 33.1).

Which open models work for Bemba?

Windy Translate lists 4 open translation models into Bemba and 5 from Bemba. All of them can be downloaded from Hugging Face and run on your own hardware. By licence: 9 Apache-2.0 (terms and attribution text: licensing). 9 of them have a screening score on FLORES-200. The best scored into Bemba is translate-en-bem (chrF++ 31.9, Usable). The best scored from Bemba is translate-bem-en (chrF++ 33.1, Usable).

With English. English to Bemba: translate-en-bem (chrF++ 31.9, Usable). Bemba to English: translate-bem-en (chrF++ 33.1, Usable). To translate between Bemba and a language with no direct model, you can go through English with these two.

By band, the scored Bemba models are 2 Usable and 7 Limited (Excellent is chrF++ 60 or more, Good 45 to 60, Usable 30 to 45, Limited 15 to 30). A score is a 48-sentence screening of news and encyclopedia text, not a measure of accuracy, and a difference of a point or two is noise. How we measured.

Speech-to-text. No speech model here covers Bemba: there is no dedicated model, and Whisper's tokenizer does not list the language.

Each model page has a short snippet to download and run the model yourself, its licence and attribution text, and its score where there is one. Pick the best scored model for your direction, then try it on a few sentences of your own text: a screening score is a starting point, not a guarantee.

Into Bemba

From Bemba

Scores: chrF++ on FLORES-200 dev, 48 sentences, measured 2026-08-04. Method.