# How to invoke GPU Docker usage

**URL:** <https://cwl.discourse.group/t/how-to-invoke-gpu-docker-usage/974>\
**Category:** CWL Questions\
**Created:** [January 24, 2025, 5:04pm UTC](https://cwl.discourse.group/t/how-to-invoke-gpu-docker-usage/974 "2025-01-24T17:04:42Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![pennycuda](https://yyz2.discourse-cdn.com/free1/user_avatar/cwl.discourse.group/pennycuda/32/582_2.png) [@pennycuda](https://cwl.discourse.group/u/pennycuda)\
**Post date:** [January 24, 2025, 5:04pm UTC](https://cwl.discourse.group/t/how-to-invoke-gpu-docker-usage/974/1 "2025-01-24T17:04:42Z")

</div>

Hello! I am a research programmer for HuBMAP Consortium. We currently have our own cwltool fork for CUDA usage in our workflows, but we would like to use the standard cwltool version if possible. Basically, I need cwltool to say `--gpus=device0:0` to Docker. Is that possible? How do I do that? Thank you!

---

<div class="post-metadata">

**Author:** ![mrc](https://yyz2.discourse-cdn.com/free1/user_avatar/cwl.discourse.group/mrc/32/3_2.png) [@mrc](https://cwl.discourse.group/u/mrc)\
**Post date:** [January 24, 2025, 8:13pm UTC](https://cwl.discourse.group/t/how-to-invoke-gpu-docker-usage/974/2 "2025-01-24T20:13:25Z")

</div>

Hello and welcome!

Please use the [`cwltool:CudaRequirement`](https://doc.arvados.org/user/cwl/cwl-extensions.html#CUDARequirement) extension and run `cwltool` or `toil-cwl-runner` using `--enable-ext`.

Don’t forget to add the following somewhere in your CWL document:

```yaml
$namespaces:
  cwltool: "http://commonwl.org/cwltool#"

```
