German, English, French, Portuguese and Spanish to Mon-Khmer languages translation model translate-tc-bible-big-deu_eng_fra_por_spa-mkh
Open machine translation from German, English, French, Portuguese and Spanish into Mon-Khmer languages, published by Windstorm Labs on Hugging Face. This model covers several target languages; see the snippet for the target token.
- Task
- Translation: German, English, French, Portuguese and Spanish → Mon-Khmer languages
- Languages
- German de, English en, French fr, Portuguese pt, Spanish es, Mon-Khmer languages mkh
- Lineage
- OPUS-MT tc-bible
- Format
- transformers (Marian), safetensors
- Hugging Face
- WindyTranslate/translate-tc-bible-big-deu_eng_fra_por_spa-mkh
- Last updated on HF
- 2026-08-04
Quality
Not yet scored. FLORES-200 has no test set for German, English, French, Portuguese and Spanish → Mon-Khmer languages, so this pair is outside the screening run.
How to run
pip install transformers sentencepiece torchfrom transformers import AutoTokenizer, AutoModelForSeq2SeqLM
repo = "WindyTranslate/translate-tc-bible-big-deu_eng_fra_por_spa-mkh"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
# Several target languages: start each input with a target token such as >>xxx<<.
# The valid tokens are listed in the upstream model card.
batch = tok([">>xxx<< <German, English, French, Portuguese and Spanish 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-tc-bible-big-deu_eng_fra_por_spa-mkh
- Upstream licence
- Apache-2.0
- Attribution
- Based on OPUS-MT model Helsinki-NLP/opus-mt-tc-bible-big-deu_eng_fra_por_spa-mkh 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:
This product uses the model WindyTranslate/translate-tc-bible-big-deu_eng_fra_por_spa-mkh
(https://huggingface.co/WindyTranslate/translate-tc-bible-big-deu_eng_fra_por_spa-mkh), published by Windstorm Labs (https://windstormlabs.com).
Based on OPUS-MT model Helsinki-NLP/opus-mt-tc-bible-big-deu_eng_fra_por_spa-mkh 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 |
|---|---|---|
| German, English, French, Portuguese and Spanish to Mon-Khmer languages translate-tcbig-bible_deu_eng_fra_por_spa-mkh | not in FLORES-200 | Apache-2.0 |
| German to English translate-de-en | 65.4 Excellent | Apache-2.0 |
| German to Danish translate-de-da | 61.4 Excellent | Apache-2.0 |
| German to French translate-de-fr | 56.7 Good | Apache-2.0 |
| German to Dutch translate-de-nl | 54.8 Good | Apache-2.0 |
| German to Italian translate-de-it | 52.3 Good | Apache-2.0 |
| German to Spanish translate-tc-big-de-es | 52.3 Good | CC-BY-4.0 |
| German to Bulgarian translate-de-bg | 52.0 Good | Apache-2.0 |