Hi, I’m submitting a container for Task2a, which is designed to utilize GPU resources. When running it locally, I include the --gpus all or runtime=nvidia flags to enable GPU support, and it works as expected. However, when running the container on the platform, I encountered the following error: 400 Client Error for http+docker://localhost/v1.51/containers/create?name=9756643: Bad Request ("unknown or invalid runtime name: nvidia") I was wondering if you could clarify whether GPU support is available for submitted containers on the platform, and if there are any specific configurations required to enable it. Thank you very much for your time and assistance.

Created by rgonl
Hi @mackostya, @Zrrr1997, @vzalevskyi24, @rgonl, Please try resubmit your Docker container as of now (August 7th, 4 p.m. UCT). Thank you, The LISA 2025 Challenge Organizers
Our team encounters the same issue with the task 2b ``` 400 Client Error for http+docker://localhost/v1.51/containers/create?name=9756738: Bad Request ("unknown or invalid runtime name: nvidia") ```
We encountered the same error on our side "unknown or invalid runtime name: nvidia". Our etag for Task2a is 28744613-e638-44e3-9b1e-44df20cb7e5b
I am encountering the same issue. Is the NVIDIA's Docker toolkit installed on the testing node?

Docker Submission Error: unknown or invalid runtime page is loading…