translate-uk-pl

Part of the Windstorm Labs open model catalogue. Scores, licence and attribution for every model: https://windytranslate.com/models/translate-uk-pl

Machine translation model, uk to pl, published by Windstorm Labs.

Attribution

Derived from Helsinki-NLP/opus-mt-uk-pl, licensed apache-2.0. Windstorm Labs did not train the original model. This notice provides the attribution the licence requires.

What was changed

The weights in this repository have been modified from the original. A full fine-tune on OPUS-100 parallel corpus data.

Method	deep-finetune
Tensors modified	253 of 254
Max absolute weight delta	1.836e-02
Verified	tensor-by-tensor against the upstream original, 2026-07-26

The comparison is tensor-level rather than file-level: safetensors and PyTorch .bin containers hash differently even when the tensors inside are identical, so a file-hash mismatch would prove nothing.

Contents

A single transformers build in safetensors format, loadable directly:

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tok = AutoTokenizer.from_pretrained("WindyTranslate/translate-uk-pl")
model = AutoModelForSeq2SeqLM.from_pretrained("WindyTranslate/translate-uk-pl")

batch = tok(["<your text here>"], return_tensors="pt", padding=True)
print(tok.batch_decode(model.generate(**batch), skip_special_tokens=True))

Other builds of this pair, including CTranslate2 INT8, are not included in this repository.

Evaluation

Benchmark	chrF++	BLEU	Band
FLORES-200 dev, 48 sentences (Ukrainian → Polish)	42.31	16.65	Usable

Measured 2026-08-04; screening score, not a publication result. Bands: Excellent ≥60, Good 45–60, Usable 30–45, Limited 15–30, Not recommended <15 (chrF++).

Limitations

• A single language direction: uk to pl.
• The evaluation above is a 48-sentence screening score, not a publication result. The weights are verified to differ from the original; that is a statement about provenance.
• Quality is not monotonic with model size or with the amount of fine-tuning applied.