Windy TranslateJapanese to Turkish translation model translate-ja-tr
Open machine translation from Japanese into Turkish, published by Windstorm Labs on Hugging Face.
- Task
- Translation: Japanese → Turkish
- Languages
- Japanese ja, Turkish tr
- Lineage
- OPUS-MT
- Format
- transformers (Marian), safetensors
- Hugging Face
- WindyTranslate/translate-ja-tr
- Last updated on HF
- 2026-08-01
Quality
| Benchmark | chrF++ | BLEU | Band | Measured |
|---|---|---|---|---|
| FLORES-200 dev, 48 sentences | 29.97 | 5.60 | 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
- 月曜日にスタンフォード大学医学部の科学者たちは、細胞を種類別に分類できる新しい診断ツールを発明したと発表しました。それは標準的なインクジェットプリンタで印刷して製造できる小型チップであり、原価は1枚あたり1円ほどす。…
- Reference
- Stanford Üniversitesi Tıp Fakültesi’nde görev alan bilim insanları, pazartesi günü hücreleri cinslerine göre s…
- This model
- Stanford Tıp Fakültesi'ndeki bilim adamları, bu yeni teşhis araçlarının Pazartesi günü üretildiğini açıkladıla
How to run
Shell
pip install transformers sentencepiece torchPython
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
repo = "WindyTranslate/translate-ja-tr"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
batch = tok(["<Japanese 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-ja-tr
- Upstream licence
- Apache-2.0
- Attribution
- Based on OPUS-MT model Helsinki-NLP/opus-mt-ja-tr 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-ja-tr
(https://huggingface.co/WindyTranslate/translate-ja-tr), published by Windstorm Labs (https://windstormlabs.com).
Based on OPUS-MT model Helsinki-NLP/opus-mt-ja-tr 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 |
|---|---|---|
| English to Turkish translate-tc-big-en-tr | 61.5 Excellent | CC-BY-4.0 |
| Bulgarian to Turkish translate-bg-tr | 47.1 Good | Apache-2.0 |
| Japanese to English translate-ja-en | 44.7 Usable | Apache-2.0 |
| Finnish to Turkish translate-fi-tr | 41.7 Usable | Apache-2.0 |
| Japanese to Portuguese translate-ja-pt | 39.8 Usable | Apache-2.0 |
| Japanese to Italian translate-ja-it | 37.9 Usable | Apache-2.0 |
| Ukrainian to Turkish translate-uk-tr | 36.5 Usable | Apache-2.0 |
| Japanese to Dutch translate-ja-nl | 36.4 Usable | Apache-2.0 |