Windy TranslateFrench to Bemba translation model translate-fr-bem
Open machine translation from French into Bemba, published by Windstorm Labs on Hugging Face.
- Task
- Translation: French → Bemba
- Languages
- French fr, Bemba bem
- Lineage
- OPUS-MT
- Format
- transformers (Marian), safetensors
- Hugging Face
- WindyTranslate/translate-fr-bem
- Last updated on HF
- 2026-08-04
Quality
| Benchmark | chrF++ | BLEU | Band | Measured |
|---|---|---|---|---|
| FLORES-200 dev, 48 sentences | 27.35 | 3.81 | Limited | 2026-08-04 |
A screening score on 48 sentences, not a publication result. Bands: Excellent ≥ 60, Good 45–60, Usable 30–45, Limited 15–30, Not recommended < 15. Method and caveats.
Recorded output (sentence 1, first 110 characters)
- Source
- Des scientifiques de l’école de médecine de l’université de Stanford ont annoncé ce lundi la création d'un nou…
- Reference
- Pali Cimo, abasayanshi ukufuma pe sukulu lya Stanford University School of Medicine babilishe ukupangwa kwa ka…
- This model
- Basayantisti abasambilila sana pa fya kundapa amalwele ya kuli bongobongo ya muntu, ababomba pa yuniversiti ya
How to run
Shell
pip install transformers sentencepiece torchPython
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
repo = "WindyTranslate/translate-fr-bem"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
batch = tok(["<French text>"], return_tensors="pt", padding=True)
print(tok.batch_decode(model.generate(**batch), skip_special_tokens=True)[0])Tested: Pattern run on 2026-10-01 against WindyTranslate/translate-tc-big-en-fr with transformers 4.57.6 and 5.18.0 on CPU ("Où se trouve la gare ?"); multi-target token prefix run against translate-tc-bible-big-mul-deu_eng_nld. Not run on every model individually.
Licence and attribution
- Licence
- Apache-2.0
- Upstream
- Helsinki-NLP/opus-mt-fr-bem
- Upstream licence
- Apache-2.0
- Attribution
- Based on OPUS-MT model Helsinki-NLP/opus-mt-fr-bem by Helsinki-NLP (Language Technology Research Group, University of Helsinki). Licensed Apache-2.0.
Copy this into your NOTICE file (or equivalent) when you ship the model:
NOTICE
This product uses the model WindyTranslate/translate-fr-bem
(https://huggingface.co/WindyTranslate/translate-fr-bem), published by Windstorm Labs (https://windstormlabs.com).
Based on OPUS-MT model Helsinki-NLP/opus-mt-fr-bem by Helsinki-NLP (Language Technology Research Group, University of Helsinki). Licensed Apache-2.0.
Windstorm Labs modified the upstream weights; the model card describes what was changed.
Licensed under the Apache License, Version 2.0: https://www.apache.org/licenses/LICENSE-2.0This is a convenience, not legal advice. See Licensing.
Related models
| Model | chrF++ (FLORES-200 dev) | Licence |
|---|---|---|
| Bemba to French translate-bem-fr | 25.2 Limited | Apache-2.0 |
| French to English translate-tc-big-fr-en | 66.2 Excellent | CC-BY-4.0 |
| French to English translate-fr-en | 63.1 Excellent | Apache-2.0 |
| French to Catalan translate-fr-ca | 55.4 Good | Apache-2.0 |
| French to Romanian translate-fr-ro | 54.3 Good | Apache-2.0 |
| French to German translate-fr-de | 51.8 Good | Apache-2.0 |
| French to Spanish translate-fr-es | 51.4 Good | Apache-2.0 |
| French to Indonesian translate-fr-id | 51.2 Good | Apache-2.0 |