Helsinki-NLP/OPUS-MT-train

Training open neural machine translation models

54
/ 100
Established

This project helps machine translation researchers and developers train custom neural machine translation (NMT) models. You provide source and target language text data, and it outputs a trained NMT model capable of translating between those languages. It is designed for those building and evaluating new translation systems.

403 stars.

Use this if you need to train your own neural machine translation models using OPUS data and the MarianNMT framework, particularly on a high-performance computing cluster.

Not ideal if you simply need to use pre-trained translation models or are looking for a user-friendly, out-of-the-box translation application.

machine-translation natural-language-processing computational-linguistics language-technology AI-research
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

403

Forks

46

Language

Makefile

License

MIT

Last pushed

Jan 17, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Helsinki-NLP/OPUS-MT-train"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.