Models
MODELHugging Face
5/5

Sunbird/translate-nllb-1.3b-salt

Lugbara(lgg)MTfacebook/nllb-200-1.3BEncoder-Decoder (NLLB fine-tune)
View Source

Model Info

Language
Lugbara (lgg)
Architecture
Encoder-Decoder (NLLB fine-tune)
Institution
Akera, Benjamin et al. (Sunbird AI / Makerere University AI Lab)
Last Updated
5 Jun 2026

Intended Use Cases

MT (English ↔ Lugbara, Luganda, Acholi, Runyankole, Ateso)

License & Access

SPDX identifierCC-BY-4.0

Open — attribution required; commercial use allowed.

Provenance

Fetch Timestamp
5 Jun 2026
Submitter
Akera, Benjamin et al. (Sunbird AI / Makerere University AI Lab)

Full Record

21 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
lgg
Model Name
Sunbird/translate-nllb-1.3b-salt
Where is it hosted
HuggingFace
Link
https://huggingface.co/Sunbird/translate-nllb-1.3b-salt
Year
2022
Creator/Author(s)
Akera, Benjamin et al. (Sunbird AI / Makerere University AI Lab)
Paper Reference
Akera et al. (2022) — Machine Translation for African Languages: Community Creation of Datasets and Models in Uganda. AfricaNLP Workshop, ICLR 2022.
NLP Task
MT
Architecture
Encoder-Decoder (NLLB fine-tune)
Base Model
facebook/nllb-200-1.3B
Fine-tuned?
Yes
License
CC-BY-4.0
Open Access
RestrictedderivedHugging Face: gated repository
Has Model Card
Yes
SOTA Tasks
MT (English ↔ Lugbara, Luganda, Acholi, Runyankole, Ateso)
Verified By
Not stated
Date Indexed
2026-06-05
Comment
Lugbara (lgg) is one of 5 Ugandan languages in the SALT MT model. Fine-tuned from NLLB-200-1.3B on the SALT parallel corpus. Also available as 4-bit (translate-…

Readiness (LRS) (4/7)

model id
sunbird-translate-nllb-1-3b-salt-lgg
Model Type
LLMderivedHugging Face card: pipeline_tag
Training Data Sources
SALT dataset + FLORES-200 + MT560 + LAFAND-MT + back-translated news
Benchmark Used
Not stated
Benchmark Score
Not stated
Benchmark Date
Not stated
Status
Activederivedupdated within 3 years

Readiness Score Breakdown

5/5
Data Coverage94%
Model Availability100%
Benchmarks100%
Documentation77%
Recency100%

Representativeness

Gender Balance
Regional Distribution
Dialect Coverage
Inclusivity
Relations Explorer link

Also Available In

This is the same multilingual corpus, sliced for other languages the platform indexes it under.

Related Models