# Translation architectures fine-tunable on a new language

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslation-architectures-fine-tunable-on-a-new-language%2F57206%3E%5C
**Category:** Models\
**Created:** [October 3, 2023, 8:55am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslation-architectures-fine-tunable-on-a-new-language%2F57206 "2023-10-03T08:55:16Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![molokanov](https://avatars.discourse-cdn.com/v4/letter/m/3e96dc/32.png) [@molokanov](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fmolokanov)%5C
**Post date:** [October 3, 2023, 8:55am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslation-architectures-fine-tunable-on-a-new-language%2F57206%2F1 "2023-10-03T08:55:16Z")

</div>

In [this tutorial](https://huggingface.co/docs/transformers/tasks/translation), several Huggingface model architectures for machine translation are listed (BART, T5, mT5, Fairseq, etc.). In order to use any of them, I should initialize it with `from_pretrained` method which means that, without fine-tuning, I can access only those languages it is pretrained on. However, I would like also to use some of these architectures for other languages that are out of the model’s scope.  
Which of them are fine-tunable on new languages?
