# Running Ray Docker image on M1/M2 Macs (arm)

**URL:** <https://discuss.ray.io/t/running-ray-docker-image-on-m1-m2-macs-arm/9236>\
**Category:** Kubernetes\
**Created:** [February 3, 2023, 3:28pm UTC](https://discuss.ray.io/t/running-ray-docker-image-on-m1-m2-macs-arm/9236 "2023-02-03T15:28:07Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![dirtyValera](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/dirtyvalera/32/1575_2.png) [@dirtyValera](https://discuss.ray.io/u/dirtyValera)\
**Post date:** [February 3, 2023, 3:28pm UTC](https://discuss.ray.io/t/running-ray-docker-image-on-m1-m2-macs-arm/9236/1 "2023-02-03T15:28:07Z")

</div>

I’m planning to upgrade my laptop to run Ray with minikube, which uses Docker. From what I see, all images are built for amd64 architecture. Does anyone have experience running these images on M1/M2 Macbooks (they have arm64 procs)? Are there any problems?

---

<div class="post-metadata">

**Author:** ![architkulkarni](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/architkulkarni/32/10_2.png) [@architkulkarni](https://discuss.ray.io/u/architkulkarni)\
**Post date:** [February 8, 2023, 12:32am UTC](https://discuss.ray.io/t/running-ray-docker-image-on-m1-m2-macs-arm/9236/2 "2023-02-08T00:32:46Z")

</div>

As of recently, it looks like we have arm64 images: [[docker/arm64] Build docker images (with suffix) for arm64 by krfricke · Pull Request #31861 · ray-project/ray · GitHub](https://github.com/ray-project/ray/pull/31861)

Can you try using the nightly images? Please let us know if you run into any issues!

---

<div class="post-metadata">

**Author:** ![teocns](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/teocns/32/6121_2.png) [@teocns](https://discuss.ray.io/u/teocns)\
**Post date:** [May 12, 2024, 9:44am UTC](https://discuss.ray.io/t/running-ray-docker-image-on-m1-m2-macs-arm/9236/3 "2024-05-12T09:44:48Z")

</div>

Hi @architkulkarni unfortunately arm64 images won’t fit in cases where running vllm…  
Currently facing issues running x86 image on M2 chip, facing the famous error:

```auto
[Bug] core_worker.cc:451: Failed to register worker 01000000ffffffffffffffffffffffffffffffffffffffffffffffff to Raylet. IOError: [RayletClient]

```

which doesn’t occur when the image targets arm64.  
Otherwise, do you have any particular setup in mind?