Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · Helsinki-NLP

opus-mt-tr-en

opus-mt-tr-en is a MarianMT Turkish-to-English translation model from Helsinki-NLP with publisher-reported OPUS benchmarks.

Publisher
Helsinki-NLP
Task
translation
Model type
marian
License
apache-2.0
Library
transformers
Publication status
Accepted · not indexed

Model overview

opus-mt-tr-en is published by Helsinki-NLP as a MarianMT translation model for Turkish to English. The captured configuration identifies MarianMTModel with model type marian. According to the model card, it uses OPUS data with normalization and SentencePiece pre-processing and reports BLEU and chr-F scores.

Recorded capabilities

Turkish-to-English translation

Hub data records the translation pipeline with tr and en tags, and the model card states Turkish source and English target.

MarianMT Transformers record

Captured configuration records MarianMTModel with model type marian and Transformers support.

Publisher-reported OPUS benchmarks

According to the model card, the publisher reports BLEU and chr-F figures on OPUS-derived newsdev, newstest, and Tatoeba sets.

Apache-2.0 record

Card data records apache-2.0.

Use cases in the source record

  • Turkish-to-English text translation using the documented OPUS-trained MarianMT setup.
  • Transformers translation workflows, including endpoints-compatible and snapshot hf-inference use with refreshed availability.

Limitations and unknowns

  • No parameter count or tokenizer vocabulary detail was extracted from this record.
  • Benchmark figures are publisher-reported model-card scores, not independently verified measurements.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: Helsinki-NLP/opus-mt-tr-en

Captured: Unknown. Processed: 2026-09-07T19:34:31.519008+00:00.

opus-mt-tr-en source languages: tr target languages: en OPUS readme: tr-en dataset: opus model: transformer-align pre-processing: normalization + SentencePiece download original weights: opus-2020-01-16.zip test set translations: opus-2020-01-16.test.txt test set scores: opus-2020-01-16.eval.txt Benchmarks testset BLEU chr-F newsdev2016-entr.tr.en 27.6 0.548 newstest2016-entr.tr.en 25.2 0.532 newstest2017-entr.tr.en 24.7 0.530 newstest2018-entr.tr.en 27.0 0.547 Tatoeba.tr.en 63.5 0.760

F001F002F003F004F005F006F007F008F009F010