Multiple Cuda
Its possible to install multiple Cuda
Create symlink /usr/local/cuda to default version
$ cd /usr/local
$ sudo rm cuda
$ sudo ln -s cuda-11.8 cuda
By default, through environment variables, the system will use the version of CUDA that the symlink /usr/local/cuda points to, and this symlink is updated when you install the new version in the previous step. If you have a certain version that you want to use as the default, update the symlink to point to that version’s installation.
4. Add each CUDA lib directory to LD_LIBRARY_PATH in order
export PATH="/usr/local/cuda-11.8/bin:$PATH"
export LD_LIBRARY_PATH="/usr/local/cuda-11.8/lib64:$LD_LIBRARY_PATH"
source .bashrc
To proceed, we must install Pytorch and TensorFlow. We can choose to install using either pip or conda environment. If you prefer to avoid manual installation of CuDNN, it would be advisable to use anaconda. Anaconda will take care of installing CuDNN for you and prevent unnecessary complications. I will be utilizing anaconda to install both GPU versions of Pytorch and TensorFlow.
https://medium.com/@peterjussi/multicuda-multiple-versions-of-cuda-on-one-machine-4b6ccda6faae