---
license: cc-by-4.0
library_name: transformers
pipeline_tag: translation
tags:
  - translation
  - marian
  - cameroon
  - low-resource
  - french-pivot
language:
  - fr
---

# français → Bulu — French-pivot MT (fp32)

Original fp32 **MarianMT** translation model for Cameroonian languages, part of the
[Lingo / NativeAI](https://lingo.cm) language-preservation project (first released
September 2024).

Translation runs through **French as a pivot** (e.g. Ghomálá' → français → Ewondo),
so each language only needs an `X→français` and a `français→X` model. The models were
trained on a corpus we compiled by hand — scripture as the aligned backbone plus
gathered books, pamphlets and other written material — normalised under the **AGLC**
alphabet (Alphabet Général des Langues Camerounaises). Background: the
[research log](https://lingo.cm/blog).

## Compressed serving bundle
For deployment we serve **int8 [CTranslate2](https://github.com/OpenNMT/CTranslate2)**
conversions (~3.8× smaller, ~6× faster on CPU), bundled at
[`flagship-ai/cameroon-int8`](https://huggingface.co/flagship-ai/cameroon-int8).
These fp32 weights are the originals / reference.

## Usage
```python
from transformers import MarianMTModel, MarianTokenizer

model = MarianMTModel.from_pretrained("flagship-ai/francais-bulu")
# Shared SentencePiece tokenizer ships in the int8 bundle subfolders:
tok = MarianTokenizer.from_pretrained("flagship-ai/cameroon-int8", subfolder="francais-bulu")

ids = tok("Bonjour, comment vas-tu ?", return_tensors="pt")
print(tok.decode(model.generate(**ids)[0], skip_special_tokens=True))
```

## Limitations
Trained on a small, formal-leaning corpus, so it is strongest on everyday/simple
sentences and weaker on modern or technical vocabulary. It is an early, open
baseline — an ongoing voice-data collection project aims to broaden coverage with
real spoken contributions.

— Open models for Cameroonian languages · [lingo.cm](https://lingo.cm)