Open bigfordy1983 opened 6 days ago
Hi. Same with #253, I think it's a cuDNN
& CUDA
& torch
version Incompatibility problem.
Can you check CUDA & torch version that you're using?
python -c "import torch; print(f'torch: {torch.__version__}'); print(f'CUDA: {torch.version.cuda}')"
The project uses CUDA 12.4 by default. If it says
torch: 2.4.1+cu124
CUDA: 12.4
And you're still having the problem,
The recommended solution is Perfview's solution, which is to manually put the dll files corresponding to your CUDA version into the system32 folder. ( cuBLAS.and.cuDNN_CUDA12_win_v1.7z
file if you're using CUDA 12.x )
Same isssue here, Perfview's solution fixed well
Which OS are you using?
Cmd Log is thus:
Use "faster-whisper" implementation Device "cuda" is detected Running on local URL: http://127.0.0.1:7860
To create a public link, set
share=True
inlaunch()
. Could not locate cudnn_ops_infer64_8.dll. Please make sure it is in your library path! "launching the app"Suspect it isn't detecting CUDA. It is installed.