I recently purchased rtx5090, and since tensorflow does not support cuda version 12.8 of rtx5090 yet, I cannot do deep learning via faceswap. If tensorflow supports cuda version 12.8 or higher in the future, I wonder if faceswap deep learning training will be possible using rtx5090. I would like to know if the faceswap development team plans to support the rtx50 series in the future. And I would like to know if the latest high-performance graphics cards such as rtx5090 require an Ubuntu environment such as wsl2 rather than windows for faceswap deep learning tasks. And I would also like to know what benefits there are for supporting faceswap.
Deep learning does not work with the RTX50 series with CUDA version 12.8 or higher.
Forum rules
Read the FAQs and search the forum before posting a new topic.
This forum is for reporting errors with the Training process. If you want to get tips, or better understand the Training process, then you should look in the Training Discussion forum.
Please mark any answers that fixed your problems so others can find the solutions.
Re: Deep learning does not work with the RTX50 series with CUDA version 12.8 or higher.
We do, but I have not had time to work on this.
Keep an eye on this thread as it is where the update will happen:
viewtopic.php?t=3010
My word is final