# Zero-shot finetuning a model for translation

**URL:** <https://discuss.huggingface.co/t/zero-shot-finetuning-a-model-for-translation/116290>\
**Category:** Beginners\
**Created:** [November 7, 2024, 9:39pm UTC](https://discuss.huggingface.co/t/zero-shot-finetuning-a-model-for-translation/116290 "2024-11-07T21:39:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ButifarraNegra](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/butifarranegra/32/34181_2.png) [@ButifarraNegra](https://discuss.huggingface.co/u/ButifarraNegra)\
**Post date:** [November 7, 2024, 9:39pm UTC](https://discuss.huggingface.co/t/zero-shot-finetuning-a-model-for-translation/116290/1 "2024-11-07T21:39:39Z")

</div>

Hey. I’m quite lost on this matter. Could somebody please clarify if it’s possible to add a new language to a model like NLLB or mT5 for translation using only monolingual data? Do I always need to finetune using data with a source and a target language (that is, parallel data)? Sorry for the ignorance. Thanks.