Skip to content

[Feature] GPU (CUDA) inference as a cuda extra #164

Open

Description

@b-re-w

The engine runs on CPU; Windows and Linux install CPU torch wheels (#139). Add a cuda extra (CUDA torch wheels) and a device option so the engine can run on an NVIDIA GPU (SPEC S1.11; PROJECT §7 M4).

Done when: uv sync --extra cuda installs CUDA torch and the engine runs on cuda with the same tests passing on a GPU runner.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type

    Projects

    No projects

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions