# What is the difference between transformers and huggingface\_hub libraries?

**URL:** <https://discuss.huggingface.co/t/what-is-the-difference-between-transformers-and-huggingface-hub-libraries/72587>\
**Category:** Beginners\
**Created:** [February 9, 2024, 8:19pm UTC](https://discuss.huggingface.co/t/what-is-the-difference-between-transformers-and-huggingface-hub-libraries/72587 "2024-02-09T20:19:09Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![BrokenSoul](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/brokensoul/32/29609_2.png) [@BrokenSoul](https://discuss.huggingface.co/u/BrokenSoul)\
**Post date:** [February 9, 2024, 8:19pm UTC](https://discuss.huggingface.co/t/what-is-the-difference-between-transformers-and-huggingface-hub-libraries/72587/1 "2024-02-09T20:19:09Z")

</div>

I’ve used transformers that allow download large language models and huggingface\_hub that only allow to download models less than 10GB.

> The model microsoft/Orca-2-13b is too large to be loaded automatically (52GB \> 10GB). Please use Spaces

  

Does huggingface\_hub allow to excess the limit if I have a pro subscription or something like that?

When to use huggingface\_hub and when transformers?, 'cause transformers allow me donwload llama 2, but in huggingface\_hub I get the message

> Bad request:  
> Model requires a Pro subscription; check out [Hugging Face – Pricing](http://hf.co/pricing) to learn more. Make sure to include your HF token in your query.

Thanks.

---

<div class="post-metadata">

**Author:** ![nielsr](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/nielsr/32/39617_2.png) [@nielsr](https://discuss.huggingface.co/u/nielsr)\
**Post date:** [February 10, 2024, 11:58am UTC](https://discuss.huggingface.co/t/what-is-the-difference-between-transformers-and-huggingface-hub-libraries/72587/2 "2024-02-10T11:58:54Z")

</div>

Hi,

Transformers is a library that contains various state-of-the-art machine learning models, as well as a Trainer API which can be used to train models.

Huggingface\_hub is a library to programmatically integrate with the hub. It allows you to programmatically list and filter models, get metadata, create pull requests, download/upload files, etc.

What’s your use case? Would you like to run a model, or only download the weights?