NNFlowVector is free to use for producing regular motion vectors, and no license is required to run it (just download and install it using the included instructions). To be able to use the smart vector output feature, you have to obtain a license (either buy one or request a trial license first).

Please do try the software on your own system before ordering a license using the shop.

If you do wish to test out any of the softwares without limitations, you can request a free trial license using this page.
By downloading our software you are accepting our end user license agreement.

If you are unsure of what CUDA version of the different builds that you should download and run, please have a look at our new CUDA compatibility chart page. If you are unsure of how to install the plugins, please read our installation instructions page.

NNFlowVector – v2.2.1

  CUDA11.8.0* MPS** CPU only
NVIDIA cuDNN v8.4.1*
NVIDIA driver 520.61+ on Linux
520.06+ on Windows
Nuke15.0
(Nuke Indie
compatible)


Nuke14.1
(Nuke Indie
compatible)

 
* This combination of CUDA & cuDNN matches what Nuke 14.1/Nuke15.0 is built against.
** MPS is Apple’s GPU acceleration on Macs called “Metal Performance Shaders” and is available on
 modern Macs with native Apple Silicon using the M series of CPUs (currently all M1, M2 and M3 variants).
The current implementation is not featuring great acceleration due to a lot of the tensor functions not being
MPS supported in the current PyTorch implementation that comes bundled with Nuke. This is hopefully
fixed in the next point release of Nuke.
  CUDA11.8.0 CUDA11.1.1* CPU only
NVIDIA cuDNN v8.4.1 v8.4.1*
NVIDIA driver 520.61+ on Linux
520.06+ on Windows
455.32+ on Linux
456.81+ on Windows
Nuke14.0
(Nuke Indie
compatible)


* This combination of CUDA & cuDNN matches what Nuke 14.0 is built against.
  CUDA11.8.0 CUDA11.2.0 CUDA10.1.243* CPU only
NVIDIA cuDNN v8.4.1 v8.1.0 v8.0.5*
NVIDIA driver 520.61+ on Linux
520.06+ on Windows
460.27+ on Linux
460.89+ on Windows
418.87+ on Linux
426.00+ on Windows
Nuke13.2
(Nuke Indie
compatible)



Nuke13.1
(Nuke Indie
compatible)



Nuke13.0
(Nuke Indie
compatible)



Nuke12.2
(Nuke Indie
compatible)


Nuke12.1

Nuke12.0

For instant Ada Lovelace/RTX40xx compatibility, please use the CUDA11.8 builds.
For instant Ampere/RTX30xx compatibility, please use the CUDA11.2 (or CUDA11.8) builds.
But for maximum compatibility, please read below: * This combination of CUDA & cuDNN matches what Nuke 13.x is built against. This means that if you do need NNFlowVector to work well together with Foundry AIR nodes (such as CopyCat) in the same Nuke scripts, you have to use this version. If you got a compute capability 8.0, 8.6, 8.9 or 9.0 GPU (i.e. Ampere/Ada Lovelace architecture, for example RTX3080), you will then need to wait for the plugin to auto JIT-compile the kernels for that GPU the first time it’s used. Please see the section about “environment variables” in the documentation to read about how to increase the CUDA_CACHE_MAXSIZE to successfully use this version with modern GPUs.