mirror of
https://github.com/invoke-ai/InvokeAI
synced 2024-08-30 20:32:17 +00:00
Update 030_INSTALL_CUDA_AND_ROCM.md
This commit is contained in:
parent
cfc4caf231
commit
40a0b2c366
@ -70,12 +70,13 @@ Runtime Site](https://developer.nvidia.com/nvidia-container-runtime)
|
||||
9. Login or create an Account.
|
||||
10. Choose the newer version of Cudnnn but __read__ the Important Note __Important Note:__
|
||||
There is two version 11x or 12x for the differents architectures(Turing,Maxwell Etc...) of GPUs.
|
||||
You can find which version you should download here note that it's based on your graphics card architecture so for exemple a 4090 use the Ada Lovelace Architecture etc.: __https://docs.nvidia.com/deeplearning/cudnn/support-matrix/index.html__
|
||||
11. Log-in Or Create An account
|
||||
12. Download the latest version and extract it from the download location
|
||||
13. Find the bin folder E\cudnn-windows-x86_64-__Whatever Version__\bin
|
||||
14. Copy and paste the 7 .dll files into YourInvokeFolderHere\\.venv\Lib\site-packages\torch\lib **Make sure to copy, and not move the files**
|
||||
15. If prompted, replace any existing files
|
||||
You can find which version you should download on the link below. note that it's based on your graphics card architecture so for exemple a 4090 use the Ada Lovelace Architecture etc.
|
||||
__https://docs.nvidia.com/deeplearning/cudnn/support-matrix/index.html__
|
||||
12. Log-in Or Create An account
|
||||
13. Download the latest version and extract it from the download location
|
||||
14. Find the bin folder E\cudnn-windows-x86_64-__Whatever Version__\bin
|
||||
15. Copy and paste the 7 .dll files into YourInvokeFolderHere\\.venv\Lib\site-packages\torch\lib **Make sure to copy, and not move the files**
|
||||
16. If prompted, replace any existing files
|
||||
|
||||
**Notes:**
|
||||
* If no change is seen or any issues are encountered, follow the same steps as above and paste the torch/lib backup folder you made earlier and replace it. If you didn't make a backup, you can also uninstall and reinstall torch through the command line to repair this folder.
|
||||
|
Loading…
Reference in New Issue
Block a user