# MultiGPU training in Code Repo & Jupyter Workspace

**URL:** <https://community.palantir.com/t/multigpu-training-in-code-repo-jupyter-workspace/5965>\
**Category:** Ask the Community\
**Tags:** code-repositories, modeling, code-workspaces\
**Created:** [February 5, 2026, 11:16am UTC](https://community.palantir.com/t/multigpu-training-in-code-repo-jupyter-workspace/5965 "2026-02-05T11:16:46Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![9eab477d6eb76c06c4f6](https://yyz1.discourse-cdn.com/palantir/user_avatar/community.palantir.com/9eab477d6eb76c06c4f6/32/7919_2.png) [@9eab477d6eb76c06c4f6](https://community.palantir.com/u/9eab477d6eb76c06c4f6)\
**Post date:** [February 5, 2026, 11:16am UTC](https://community.palantir.com/t/multigpu-training-in-code-repo-jupyter-workspace/5965/1 "2026-02-05T11:16:46Z")

</div>

Hello,

I am trying to run Multi-GPU training in Foundry but I would like some expertise from the community:

- Spark with profile=[“DYNAMIC\_ALLOCATION\_ENABLED”,“DRIVER\_GPU\_ENABLED”] → One GPU get picked up from RQ and is “activated” during build but from what I understand the Driver shouldn’t be the one used for this task in a Spark infra.

 ![image](https://canada1.discourse-cdn.com/palantir/original/2X/7/765f995b9ea7d29f228e4b5e76e29ec7f2721543.png)

- Spark with profile=[@configure(profile=[“DYNAMIC\_ALLOCATION\_ENABLED”, “EXECUTOR\_GPU\_ENABLED”])] → I believe that’s working as expected.

 ![image](https://canada1.discourse-cdn.com/palantir/original/2X/4/4872e1bd07e7de92b133243fbbfba2c2ab7dba01.png)

I understand that Palantir recommends the usage of Lightweight for this usage however no arguments available - like gpu\_count - to set GPU to 2 ?

With Lightweight: @lightweight(gpu\_type=“NVIDIA\_A10G”), only One GPU get picked up from RQ.

As well, I am not sure how to enable such capacity on Code Workspace / Jupyter Notebook ?

Best regards,