Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · Helsinki-NLP

opus-mt-bn-en

opus-mt-bn-en is a Marian Bengali-to-English translation model from Helsinki-NLP. According to the model card, it reports a Tatoeba BLEU of 49.7.

Publisher
Helsinki-NLP
Task
translation
Model type
marian
License
apache-2.0
Library
transformers
Publication status
Approved for indexing

Model overview

opus-mt-bn-en is published by Helsinki-NLP as a translation model. The captured configuration identifies MarianMTModel with model type marian. According to the model card, it covers the Bengali-to-English pair with transformer-align modeling and SentencePiece preprocessing.

Recorded capabilities

Bengali-to-English pair

According to the model card, Bengali is the source group and English is the target group, with non-multilingual source and target configuration.

Documented preprocessing

According to the model card, inputs use normalization with SentencePiece spm32k models.

Published Tatoeba score

According to the model card, the Tatoeba Bengali-English test reports BLEU 49.7 with chr-F 0.641.

Use cases in the source record

  • Bengali-to-English translation through the documented normalization and SentencePiece pipeline.
  • Score comparison using the card's published Tatoeba BLEU and chr-F figures.

Limitations and unknowns

  • No parameter count was extracted from this record.
  • No context-window value was extracted from this record.
  • BLEU and chr-F figures are publisher-reported benchmark claims, not independently verified.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

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

Captured: Unknown. Processed: 2026-09-07T19:34:31.249800+00:00.

ben-eng source group: Bengali target group: English OPUS readme: ben-eng model: transformer-align source language(s): ben target language(s): eng model: transformer-align pre-processing: normalization + SentencePiece (spm32k,spm32k) download original weights: opus-2020-06-17.zip test set translations: opus-2020-06-17.test.txt test set scores: opus-2020-06-17.eval.txt Benchmarks testset BLEU chr-F Tatoeba-test.ben.eng 49.7 0.641 System Info: hf_name: ben-eng source_languages: ben target_languages: eng opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/ben-eng/README.md original_repo: Tatoeba-Challen…

F001F002F003F004F005F006F009F010F011