---
language_details: "spa_Latn, quy_Latn, nso_Latn, fra_Latn, bos_Latn"
pipeline_tag: translation
tags:
- nllb
license: "cc-by-nc-4.0"
inference: false
---

# Description
Finetuned [facebook/nllb-200-3.3B](https://huggingface.co./facebook/nllb-200-3.3B) model to translate between Spanish ("spa_Latn") and Mapuzungún.
We support the following languages/graphemaries:
- Spanish (**spa_Latn**)
- Azümchefe (**quy_Latn**)
- Ragileo (**nso_Latn**)
- Unificado (**fra_Latn**)

# Example
```python
from transformers import NllbTokenizerFast, AutoModelForSeq2SeqLM

tokenizer = NllbTokenizerFast.from_pretrained("CenIA/nllb-200-3.3B-spa-arn")
model = AutoModelForSeq2SeqLM.from_pretrained("CenIA/nllb-200-3.3B-spa-arn")

def translate(sentence: str, translate_from="spa_Latn", translate_to="quy_Latn") -> str:
    tokenizer.src_lang = translate_from
    tokenizer.tgt_lang = translate_to

    inputs = tokenizer(sentence, return_tensors="pt")
    result = model.generate(**inputs, forced_bos_token_id=tokenizer.convert_tokens_to_ids(translate_to))
    decoded = tokenizer.batch_decode(result, skip_special_tokens=True)[0]

    return decoded

traduction = translate("Hola, ¿cómo estás?")

print(traduction)
```