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

why cant i run live #948

Open
sirpine00 opened this issue Feb 20, 2025 · 4 comments
Open

why cant i run live #948

sirpine00 opened this issue Feb 20, 2025 · 4 comments

Comments

@sirpine00
Copy link

sirpine00 commented Feb 20, 2025

Ive been trying for so long to get live to work and everything ive tried doesnt work. i am using cuda 11.8, onnxruntime-gpu 1.16.3 and python 3.10.0.

CUDA_PATH is set but CUDA wasn't able to be loaded. Please install the correct version of CUDA and cuDNN as mentioned in the GPU requirements page (https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html#requirements), make sure they're in the PATH, and that your GPU is supported.

thats what it says and im comepletely lost, i am also very new to all of this and i followed the readme file btw.

By the way, i can run the normal one (the one without cuda) but it is like 1 fps and unusable. Whenever i boot the cuda version up, the preview just doesnt load.

@lywaterman
Copy link

the same question

@daryltucker
Copy link

daryltucker commented Feb 21, 2025

It seems that when you download and install cuDNN, even though it claims it will install for CUDA 12.8 and 11.8, the 11.8 version isn't compatible.

What resolved this issue for me was to use the v8.9.7 cuDNN for CUDA 11.x zip from the NVIDIA Archive.

MD5: 7dddb764c0a608ac23e72761be4c92c0
SHA256: 5e45478efe71a96329e6c0d2a3a2f79c747c15b2a51fead4b84c89b02cbf1671

After downloading the .zip, it creates cudnn-windows-x86_64-8.9.7.29_cuda11-archive directory. If you already have cuDNN installed, you should have a directory C:\Program Files\NVIDIA\CUDNN with your cuDNN versions. I renamed the cudnn-windows-x86_64-8.9.7.29_cuda11-archive to 8.9.7.29_cuda11 and placed it within that directory.

Then, I find "Edit the system environment variables" in Windows and click "Environment Variables". From there, you can edit the User or System variable of Path to add a new path. The path for me was C:\Program Files\NVIDIA\CUDNN\8.9.7.29_cuda11\bin, and after adding it, I made sure to click "Move Up" to seat it above any other CUDNN entries.

After installing the appropriate cuDNN, setting the Path variable, and completely closing and re-opening the terminal, I was able to get this working.

Hope this was helpful.

@Xiller
Copy link

Xiller commented Feb 22, 2025

It seems that when you download and install cuDNN, even though it claims it will install for CUDA 12.8 and 11.8, the 11.8 version isn't compatible.

What resolved this issue for me was to use the v8.9.7 cuDNN for CUDA 11.x zip from the NVIDIA Archive.

MD5: 7dddb764c0a608ac23e72761be4c92c0 SHA256: 5e45478efe71a96329e6c0d2a3a2f79c747c15b2a51fead4b84c89b02cbf1671

After downloading the .zip, it creates cudnn-windows-x86_64-8.9.7.29_cuda11-archive directory. If you already have cuDNN installed, you should have a directory C:\Program Files\NVIDIA\CUDNN with your cuDNN versions. I renamed the cudnn-windows-x86_64-8.9.7.29_cuda11-archive to 8.9.7.29_cuda11 and placed it within that directory.

Then, I find "Edit the system environment variables" in Windows and click "Environment Variables". From there, you can edit the User or System variable of Path to add a new path. The path for me was C:\Program Files\NVIDIA\CUDNN\8.9.7.29_cuda11\bin, and after adding it, I made sure to click "Move Up" to seat it above any other CUDNN entries.

After installing the appropriate cuDNN, setting the Path variable, and completely closing and re-opening the terminal, I was able to get this working.

Hope this was helpful.

THANKS YES.

i tried with CuDNN 8.6 , and 8.9 .. maybe it helps someone:

The solution was for me to put it in PATH. i put it in CuDNN "Environment Variables" in system, not only in PATH . so i put 2 Positions in Systemvariable:

  1. cuDNN with: C:\Program Files\NVIDIA\CUDNN\v8.9;C:\Program Files\NVIDIA\CUDNN\v8.9\bin;C:\Program Files\NVIDIA\CUDNN\v8.9\include;C:\Program Files\NVIDIA\CUDNN\v8.9\lib;

  2. PATH with: C:\Program Files\NVIDIA\CUDNN\v8.9\bin

Thanks @daryltucker !!

@premhaske1109
Copy link

watch this tutorial properly!!
https://youtu.be/MICScLVM1x8?si=9_BylTaf9RlYKYUM

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

5 participants