On windows, how do you verify the version number of CuDNN installed?
I'm finding a lot of results when I search for the answer for Linux machines. It doesn't seem like there's a clear way to find out what version is installed for Windows.
Asked
Active
Viewed 5.5k times
4 Answers
43
Go to C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\include\
open cudnn.h
define CUDNN_MAJOR 5
define CUDNN_MINOR 1
define CUDNN_PATCHLEVEL 10
in my case its 5.1.10
Yeghia
- 510
- 5
- 8
-
1Perfect. Thank you. – Mauker Jan 29 '18 at 23:01
-
2On my machine cudnn header was installed in `C:\Program Files\cuDNN6\cuda\include` – Shital Shah Jul 05 '18 at 22:03
-
1In my case `cudnn.h` does not define the version numbers, but it rather imports them from `cudnn_version.h`, so I just had to open that file instead. – Redoman May 26 '22 at 15:10
-
This is assuming that CuDNN files have been copied to this locatoin, but they could be anywhere on the system, and the location included in the system PATH variable. – JStrahl Nov 09 '22 at 10:30
2
#cudnn version check (win10)
in my case its cuda 11.3v // u need to replace with ur verisons
more "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3\include\cudnn_version.h"
output:
Bharath Kumar
- 893
- 9
- 8
-
1you should not replace anything here, but I +1 this because the path has changed since cuda 11 and now it is under cudnn_version – prhmma Jul 04 '21 at 08:06
1
The easiest way to get CUDA installed version info is by checking the installation folder as shown below. Here CUDA version is v11.7.
Or try nvcc -V in the installed environment as shown below figure.
surendra Allam
- 61
- 8
-2
In my case you should go in: "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2\include"
and later, open cudnn_version.h
in my case it shows:
#define CUDNN_MAJOR 8
#define CUDNN_MINOR 1
#define CUDNN_PATCHLEVEL 0
finally, my version is: 8.1.0
