# How to install MPS backend on docker/podman container?

**URL:** <https://dev-discuss.pytorch.org/t/how-to-install-mps-backend-on-docker-podman-container/3015>\
**Category:** Uncategorized\
**Created:** [May 21, 2025, 5:20pm UTC](https://dev-discuss.pytorch.org/t/how-to-install-mps-backend-on-docker-podman-container/3015 "2025-05-21T17:20:02Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![pytorch1](https://avatars.discourse-cdn.com/v4/letter/p/6bbea6/32.png) [@pytorch1](https://dev-discuss.pytorch.org/u/pytorch1)\
**Post date:** [May 21, 2025, 5:20pm UTC](https://dev-discuss.pytorch.org/t/how-to-install-mps-backend-on-docker-podman-container/3015/1 "2025-05-21T17:20:02Z")

</div>

I’m trying to install pytorch with MPS backend - has anyone tried installing it on Docker?

---

<div class="post-metadata">

**Author:** ![albanD](https://yyz2.discourse-cdn.com/flex036/user_avatar/dev-discuss.pytorch.org/alband/32/9_2.png) [@albanD](https://dev-discuss.pytorch.org/u/albanD)\
**Post date:** [May 21, 2025, 8:55pm UTC](https://dev-discuss.pytorch.org/t/how-to-install-mps-backend-on-docker-podman-container/3015/2 "2025-05-21T20:55:05Z")

</div>

Hey!

Did you follow the instructions in [https://pytorch.org/get-started/locally/](https://pytorch.org/get-started/locally/) to install inside your docker image?

---

<div class="post-metadata">

**Author:** ![pytorch1](https://avatars.discourse-cdn.com/v4/letter/p/6bbea6/32.png) [@pytorch1](https://dev-discuss.pytorch.org/u/pytorch1)\
**Post date:** [August 18, 2025, 10:08am UTC](https://dev-discuss.pytorch.org/t/how-to-install-mps-backend-on-docker-podman-container/3015/3 "2025-08-18T10:08:41Z")

</div>

Thanks @albanD How can I set up PyTorch in editable mode on my Macbookso that I can make code changes locally and have those changes immediately reflected when I use the library? I want to configure my environment for contributing to PyTorch. But online tutorials are very few including docs which are not very clear. I’m ideally looking at setting it up in a dev container. Example: [GitHub - keras-team/keras-cv: Industry-strength Computer Vision workflows with Keras](https://github.com/keras-team/keras-cv) has a pretty neat dev container file making it very easy to setup.