# Ray-ml docker image for armv7 architecture?

**URL:** <https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628>\
**Category:** Ray Core\
**Created:** [January 27, 2021, 10:23pm UTC](https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628 "2021-01-27T22:23:30Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Peter\_Pirog](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/peter_pirog/32/234_2.png) [@Peter\_Pirog](https://discuss.ray.io/u/Peter_Pirog)\
**Post date:** [January 27, 2021, 10:23pm UTC](https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628/1 "2021-01-27T22:23:30Z")

</div>

I wonder if is possible to make docker images ray-ml:

[https://hub.docker.com/r/rayproject/ray-ml](https://hub.docker.com/r/rayproject/ray-ml) for architectures:

> linux/arm64,linux/arm/v7, linux/arm/v6

I tried to build from ray-ml image with:

> FROM rayproject/ray-ml  
> CMD [“/bin/bash”]

> docker buildx build --platform linux/arm/v7 --push -t peterpirogtf/ray\_tf2:armv7 .

Unfortunatelly the error:

> standard\_init\_linux.go:219: exec user process caused: exec format error

suggest (in my opinion) that there is some problem with architecture compatibility.

---

<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:** [April 14, 2023, 3:41pm UTC](https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628/2 "2023-04-14T15:41:41Z")

</div>

Bumping this up, interested in how to build Ray images on arm machines

---

<div class="post-metadata">

**Author:** ![jjyao](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/jjyao/32/1799_2.png) [@jjyao](https://discuss.ray.io/u/jjyao)\
**Post date:** [April 15, 2023, 3:18am UTC](https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628/3 "2023-04-15T03:18:23Z")

</div>

Hi @dirtyValera, we do have now: e.g, `docker pull rayproject/ray:2.3.0-py38-aarch64`

---

<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:** [April 15, 2023, 6:10am UTC](https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628/4 "2023-04-15T06:10:22Z")

</div>

@jjyao, thanks for reply! What about earlier versions, 2.0.1 specifically. I can’t find rayproject/ray:2.0.1-py38-aarch64. Is there any guide how to build it locally? Also please see similar question I asked [[arm64][Docker] How to build/download Arm64 Docker image for specific version of Ray?](https://discuss.ray.io/t/arm64-docker-how-to-build-download-arm64-docker-image-for-specific-version-of-ray/10231)

And what about ray-ml images?

---

<div class="post-metadata">

**Author:** ![jjyao](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/jjyao/32/1799_2.png) [@jjyao](https://discuss.ray.io/u/jjyao)\
**Post date:** [April 26, 2023, 6:18pm UTC](https://discuss.ray.io/t/ray-ml-docker-image-for-armv7-architecture/628/5 "2023-04-26T18:18:07Z")

</div>

Per [https://github.com/ray-project/ray/pull/31861](https://github.com/ray-project/ray/pull/31861), we don’t have arm support for ray-ml due to some dependency issue.