O

Opus Mt Tc En Pl

Developed by gsarti
This is a neural machine translation model for English to Polish based on the Transformer architecture, trained by the Helsinki NLP team using the MarianNMT library.
Downloads 243
Release Time : 3/2/2022

Model Overview

This model is specifically designed for translating English text into Polish, trained on the Tatoeba dataset, and suitable for general text translation tasks.

Model Features

High-quality translation
Achieves a BLEU score of 47.5 and a chr-F score of 0.673 on the Tatoeba test set, demonstrating excellent performance.
Trained on large-scale data
Trained using 59,742,979 English-Polish sentence pairs from the Tatoeba dataset.
Standardized preprocessing
Employs normalization and SentencePiece tokenization (spm32k) for text preprocessing.

Model Capabilities

English to Polish text translation
Handling general domain text

Use Cases

Text translation
Daily conversation translation
Translating English daily conversations into Polish
Performs well on the Tatoeba test set
Content localization
Localizing English content into Polish versions
Featured Recommended AI Models
ยฉ 2025AIbase