Models
MODELHugging Face
5/5

Helsinki-NLP/opus-mt-en-mg

Malagasy(mlg)MTMarianMTEncoder-Decoder (MarianMT)
View Source

Model Info

Language
Malagasy (mlg)
Architecture
Encoder-Decoder (MarianMT)
Institution
Helsinki-NLP (Language Technology Research Group, U. of Helsinki)
Last Updated
9 Jun 2026

Intended Use Cases

MT (EN→MG)

License & Access

SPDX identifierApache-2.0

Open — attribution required; commercial use allowed.

Provenance

Fetch Timestamp
9 Jun 2026
Submitter
Helsinki-NLP (Language Technology Research Group, U. of Helsinki)

Full Record

20 of 25 fields stated

Every column the registry tracks. Values marked derivedwere filled by a rule or read from the source's text rather than stated in its metadata; the rule or the quoted sentence is shown under the value.

Registry columns (17/18)

iso code
mlg
Model Name
Helsinki-NLP/opus-mt-en-mg
Where is it hosted
HuggingFace
Link
https://huggingface.co/Helsinki-NLP/opus-mt-en-mg
Year
2020
Creator/Author(s)
Helsinki-NLP (Language Technology Research Group, U. of Helsinki)
Paper Reference
Tiedemann, Jörg (2020). OPUS-MT. LREC 2020.
NLP Task
MT
Architecture
Encoder-Decoder (MarianMT)
Base Model
MarianMT
Fine-tuned?
Yes
License
Apache-2.0
Open Access
Open
Has Model Card
Yes
SOTA Tasks
MT (EN→MG)
Verified By
Not stated
Date Indexed
2026-06-09
Comment
English→Malagasy MT model from Helsinki-NLP OPUS-MT series. Multiple directional models available (en-mg, mg-en, fi-mg etc). Malagasy ISO 639-1: mg.

Readiness (LRS) (3/7)

model id
helsinki-nlp-opus-mt-en-mg-mlg
Model Type
MTderivedHugging Face card: pipeline_tag
Training Data Sources
OPUS parallel corpus (en-mg)
Benchmark Used
Not stated
Benchmark Score
Not stated
Benchmark Date
Not stated
Status
Not stated

Readiness Score Breakdown

5/5
Data Coverage100%
Model Availability100%
Benchmarks100%
Documentation86%
Recency100%

Representativeness

Gender Balance
Regional Distribution
Dialect Coverage
Inclusivity
Relations Explorer link

Other Copies

The same resource, also published at these locations. It is counted once.

Related Models