# About the Flax/JAX Projects category

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-flax-jax-projects-category%2F7061%3E%5C
**Category:** Flax/JAX Projects\
**Created:** [June 22, 2021, 1:09pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-flax-jax-projects-category%2F7061 "2021-06-22T13:09:07Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![sgugger](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/sgugger/32/2291_2.png) [@sgugger](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsgugger)%5C
**Post date:** [June 22, 2021, 1:09pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-flax-jax-projects-category%2F7061%2F1 "2021-06-22T13:09:07Z")

</div>

This category should be used to propose and join existing projects that make use of :jax: Flax/JAX and 🤗 Transformers. Make sure that your project proposal includes all relevant information (as listed below) for people to join/comment/improve your idea.

Feel free to also post any questions you might have about 🤗 Transformers [flax examples](https://github.com/huggingface/transformers/tree/master/examples/flax). If you found issues in one of the flax examples, please [submit an issue](https://github.com/huggingface/transformers/issues/new/choose) to the transformers repo instead.

Relevant information when proposing a :jax:/🤗 project:

- Description: What is the project about? _E.g._ Train a CLIP model in Swedish for zero-shot image classification.
- Model: What model do you propose to use? _E.g._ [openai/clip-vit-base-patch32 · Hugging Face](https://huggingface.co/openai/clip-vit-base-patch32) in Flax/JAX
- Dataset: What dataset do you want to use? _E.g._ [GitHub - google-research-datasets/wit: WIT (Wikipedia-based Image Text) Dataset is a large multimodal multilingual dataset comprising 37M+ image-text sets with 11M+ unique images across 100+ languages.](https://github.com/google-research-datasets/wit)
- Training scripts: What training scripts do you want to use? _E.g._ Training scripts will be created as part of the project
- Expected result: What do you expect the outcome of the project to be? _E.g._ A zero-shot image classification pipeline for CLIP

---

<div class="post-metadata">

**Author:** ![skylord](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/skylord/32/49255_2.png) [@skylord](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fskylord)%5C
**Post date:** [June 27, 2021, 2:05am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-flax-jax-projects-category%2F7061%2F2 "2021-06-27T02:05:50Z")

</div>

Hi - I wanted to propose a project. But the submitted projects are really interesting. Is it possible to join more than one ?

---

<div class="post-metadata">

**Author:** ![valhalla](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/valhalla/32/4179_2.png) [@valhalla](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvalhalla)%5C
**Post date:** [June 27, 2021, 6:03am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-flax-jax-projects-category%2F7061%2F3 "2021-06-27T06:03:00Z")

</div>

Yes, you can join multiple projects!

---

<div class="post-metadata">

**Author:** ![hugbump](https://avatars.discourse-cdn.com/v4/letter/h/3da27b/32.png) [@hugbump](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fhugbump)%5C
**Post date:** [July 1, 2022, 2:45pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fabout-the-flax-jax-projects-category%2F7061%2F4 "2022-07-01T14:45:01Z")

</div>

Just noticed this category, which has quite many topics already. Wow.  
A stupid question, I see in README of `JAX/Flax Examples`:

“This folder contains actively maintained examples of 🤗 Transformers using the JAX/Flax backend. Porting models and examples to JAX/Flax is an ongoing effort … In particular, these examples are all designed to run fast on Cloud TPUs, and we include step-by-step guides to getting started with Cloud TPU.”

I heard of JAX/Flax before in an nlp coursera class. My understanding is that it is a backend different from tensorflow or pytorch, but why making a big deal out of porting to JAX/Flax? What difference does it bring? I guess the difference and gain must be really big to worth the porting. Thank you very much!
