Skip to content

Support for Tensorflow on newer CUDA versions #3851

Description

@kateireed

Description

Are there plans to add support for running Tensorflow on newer CUDA versions? On Maven it looks like the most recently available dependency is tensorflow-native-cu121 from 2024. Do you expect to support newer CUDA versions or plan to phase out GPU support for Tensorflow over time?

Will this change the current api? How?
No, just make existing libraries work on newer GPUs.

Who will benefit from this enhancement?
Users with newer GPUs that aren't backwards compatible to the currently supported CUDA versions.

References

P.S. Thanks for all the work you do! We are using DJL to deploy Tensorflow, Pytorch, and ONNX acoustic detector models cross-platform for conservation bioacoustics projects all over the world, and it works great!

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions