Run `nvcc –version`
Run `nvidia-smi`
Check `cat /usr/local/cuda/version.txt`
Check `ls -l /usr/local/cuda`
In Python, run `import torch; print(torch.version.cuda)` if using PyTorch
In Python, run `import tensorflow as tf; print(tf.sysconfig.get_build_info()[“cuda_version”])` if using TensorFlow
On Windows, check the CUDA version in the NVIDIA CUDA Toolkit installation folder
On Linux, check the CUDA version with `dpkg -l | grep cuda` or `rpm -qa | grep cuda`