I installed cuda 8.0 in my ubuntu 16.04 machine and checked the cuda version using the command "nvcc --version". it shows version as 7.5!!!.How Can I be sure that it is accurate? Are there other commands that I can also use to verify my result?
-
1What's your GPU ? Did you have `cuda` previously installed? – pSoLT Jan 18 '17 at 08:48
-
I have GTX970 GPU. No I didn't installed cuda previously in my system. – Sowmya Dhanapal Jan 18 '17 at 08:54
-
How did you install `cuda` ? – pSoLT Jan 18 '17 at 09:01
-
I downloaded the package from "https://developer.nvidia.com/compute/cuda/8.0/prod/local_installers/cuda-repo-ubuntu1604-8-0-local_8.0.44-1_amd64-deb" and I followed the installation instructions. – Sowmya Dhanapal Jan 18 '17 at 09:53
-
Instead of the package, please provide a link to the **instructions** you followed. – Taro Jan 18 '17 at 09:55
-
I followed the instructions in the following link http://www.r-tutor.com/gpu-computing/cuda-installation/cuda7.5-ubuntu but instead of 7.5, I replaced with 8.0. – Sowmya Dhanapal Jan 18 '17 at 10:21
-
4What path did you install CUDA 8.0 to? What is the output of `which nvcc`? – tera Jan 18 '17 at 11:35
-
If you are looking for the actual version digits as a string you can use a mix of `nvidia-smi` to get the version, `grep` to identify the line, and `sed` to remove the unnecessary characters: `nvidia-smi | grep -o 'CUDA Version: [0-9].\.[0-9]' | sed 's/.*: //'` – Eduardo Pignatelli Feb 16 '21 at 20:24
4 Answers
[edited 2022]
For CUDA 11:
$ cat /usr/local/cuda/version.json
For cuda-8.0 on Ubuntu16.04, you should be able to read
$ cat /usr/local/cuda/version.txt
CUDA Version 8.0.44
I agree with Robert Crovella, you might need to check your PATH

- 1,919
- 13
- 13
-
6/usr/local/cuda is a symlink to the version you have installed. You can install cuda without the symlink and then this will not work. – Punnerud Feb 14 '18 at 11:29
-
2
Starting from CUDA 8.0, it's possible to have multiple CUDA versions installed. You can then activate different values for $PATH
environment variable that will present you with different CUDA version.
Command to immediately obtain the CUDA version:
$ nvcc --version | grep "release" | awk '{print $6}' | cut -c2-
You can confirm the result by checking the install status of CUDA libraries:
$ dpkg -l | grep cuda
For installing multiple versions of CUDA, you can refer to this article.

- 299
- 2
- 14

- 399
- 3
- 4
Thank you all... Previously I tried to install cuda8.0 using run file from https://developer.nvidia.com/compute/cuda/8.0/prod/local_installers/cuda_8.0.44_linux-run. After that I tried to check "nvcc --version", but it shows the following error "The program 'nvcc' is currently not installed. You can install it by typing: sudo apt-get install nvidia-cuda-toolkit". So I tried the above command. It gave the cuda7.5 version.
Later I tried to install cuda using debian package which by default contained nvcc. Now I am getting correct version.

- 471
- 1
- 4
- 9
-
3Follow the directions in the linux getting started guide. When you use a runfile installer, you must update your `PATH` environment variable. If you don't do that, you will get the message about `nvcc` not being installed. – Robert Crovella Jan 21 '17 at 04:37
It may be due to the fact that you have both v7.5 and v8.0 installed. So instead of changing path, try uninstalling v7.5 first

- 260
- 7
- 18