Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

New docker images #392

Open
martabal opened this issue Jul 3, 2024 · 0 comments
Open

New docker images #392

martabal opened this issue Jul 3, 2024 · 0 comments

Comments

@martabal
Copy link
Collaborator

martabal commented Jul 3, 2024

Hello everyone!

We are adding cuda and openvino support for GPU acceleration. Until now, if you wanted to have GPU acceleration for your Nvidia GPU, you had to install a bunch of dependencies during the runtime. Everytime you restarted your Immich container, you installed/checked the dependencies needed for the GPU acceleration. This could take several minutes, was inefficient and could lead to dependency installation issues. It's no more the case, all dependencies are now installed during the image build.

We don't want to ship a 3 GB docker image for people who don't use machine learning, so we are now maintaining 4 different versions of Immich:

  • noml: (stands for "no machine-learning") Latest version of Immich without machine-learning
  • latest: Latest version of Immich (obviously)
  • cuda: Latest version of Immich and supports cuda for GPU acceleration (Nvidia GPUs)
  • openvino: Latest version of Immich and supports openvino for GPU acceleration (Intel GPUs)

To use one of these versions, simply replace the docker tag with that version (for exemple ghcr.io/imagegenius/immich:cuda)

To use cuda or openvino versions, please follow the instructions here.

If you were using cuda for GPU acceleration, the MACHINE_LEARNING_GPU_ACCELERATION env is no longer used.

Warning

If you were using cuda, remove the cuda folder in /config/machine-learning

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant