Is a GTX 1080 compatible with this? #1088
Replies: 3 comments
-
|
you must uninstall libtorchcodec which is part of torch 2.9... libtorchcodec replaces torchaudio. you have torchaudio version conflict. |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
please test 25.11.25 including smart torch installation for your GPU |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
v25.12.18 has automatic GPU detection now. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
I am a novice when it comes to docker and command line, but I have enlisted the help of some other folks and I can't seem to get my GPU to work.
--I first rebuilt the docker container with CUDA-118 and that didn't work.
--nvidia-smi indicated that I was running CUDA 12.8 so I then rebuilt with CUDA 128
--I received similar errors, the only difference being that the error indicated the CUDA version I had installed at each attempt (118 for that attempt, 128 for the other).
--Then I edited the yaml file to enable the GPU and it still didnt work, providing the same error when I rebuilt at CUDA 118 and 128,
Included are my CUDA 128 log, my yaml file, and my attempted audiobook generation using the GPU after the CUDA 128 error (i figured it wouldnt work but maybe it gives you a clue).
cuda-128_log.txt
generation error.txt
docker-compose.yml
I tried to install locally and got a similar issue. I couldn't run it with the included 2.9 version of Torch, so I followed shadeau9's guide and the wiki. I used shadeau's uninstall command and reinstalled Cuda 128 (Torch ver. 2.7) using the provided command in the GPU wiki. Here is the error I got.
local_error_2-7.txt
Beta Was this translation helpful? Give feedback.
All reactions