To be able to install and run any of our plugins, you must first identify what OS you are running (Linux, Windows or macOS) and also what version of Nuke. When you know that, you must pick a suitable CUDA version for your needs (if applicable). If you are unsure of what CUDA version of the different builds that you should download and run, please have a look at our CUDA compatibility chart page. Also be sure to check that you have a supported NVIDIA GPU and that your currently installed driver version is modern enough. The macOS builds are accelerated using MPS instead of CUDA.
Linux
- Start by downloading the correct build (see above).
- Extract the content of the zip file to a directory of your own choice.
- Add the full path of that new directory to the NUKE_PATH environment variable.
Windows
- Start by downloading the correct build (see above).
- Extract the content of the zip file to a directory of your own choice.
- Add the full path of that new directory to the NUKE_PATH environment variable.
- If you can’t open Nuke, and it’s complaining about “The specified module could not be found”, then add the full path to the new install directory to the environment variable PATH as well. You should now be able to start Nuke and use the plugin.
macOS
- Start by downloading the correct build. You might need to right click on the Apple logo and pick “Save link as”.
- The downloaded file is a self extracting shell script archive. You need to run it from a terminal, for example:
cd /full/path/to/download/folder
sh ./NNFlowVector-v2.2.1_Nuke15.0_MPS_Mac.sh - The shell script will extract it’s content into a new folder named the same as the file (but without the “.sh” extension).
- Add the full path of that new directory to the NUKE_PATH environment variable.