---
datasets:
- AfriNLLB-train
- AfriNLLB-train-distilled
language:
- ar
- am
- af
- arz
- es
- en
- fr
- ha
- ln
- pt
- so
- sw
- wo
- yo
- zu
metrics:
- bleu
- chrf
- comet
library_name: transformers
base_model:
- facebook/nllb-200-distilled-600M
- AfriNLP/AfriNLLB-12enc-12dec-full-ft-kd
pipeline_tag: translation
license: cc-by-nc-4.0
tags:
- pruning
new_version: AfriNLP/AfriNLLB-12enc-8dec-iterative-548m-ft
---

# AfriNLLB

AfriNLLB is a series of efficient multilingual open-source models for African languages. 

## Model Description

This model is created using *middle layer pruning*. We recommend using the models created with *iterative layer pruning*. Please refer to the paper for more details.

## Citation

If you use any of AfriNLLB models, datasets, or approaches, please cite the following paper:

```bibtex
@inproceedings{moslem-etal-2026-afrinllb,
    title = "{A}fri{NLLB}: Efficient Translation Models for African Languages",
    author = "Moslem, Yasmin  and
      Wassie, Aman Kassahun  and
      Gizachew, Amanuel",
    booktitle = "Proceedings of the Seventh Workshop on African Natural Language Processing (AfricaNLP)",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://openreview.net/forum?id=hVJZNUZBur",
}
```