You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
## How to use a CUDA version that is different from PyTorch
2
+
3
+
Some features of bitsandbytes may need a newer CUDA version than regularly supported by PyTorch binaries from conda / pip. In that case you can use the following instructions to load a precompiled bitsandbytes binary that works for you.
4
+
5
+
## Installing or determining the CUDA installation
6
+
7
+
Determine the path of the CUDA version that you want to use. Common paths paths are:
8
+
```bash
9
+
/usr/local/cuda
10
+
/usr/local/cuda-XX.X
11
+
```
12
+
13
+
where XX.X is the CUDA version number.
14
+
15
+
You can also install CUDA version that you need locally with a script provided by bitsandbytes as follows:
It is best to add these lines to the `.bashrc` file to make them permanent.
46
+
47
+
If you now launch bitsandbytes with these environmental variables the PyTorch CUDA version will be overridden by the new CUDA version and a different bitsandbytes library is loaded (in this case version 117).
0 commit comments