Licences requirement for NVIDIA Pytorch dependencies

Accelerated Computing CUDA CUDA Programming and Performance
,
1

hey team!
We are planning to use the pytorch library within our organisation but there are these dependencies of the library which are listed as NVIDIA Proprietary Software. Is that something that we need to get license to use or is this open source and we can go ahead and use it within our org?

These are the libraries:
–nvidia-cublas-cu12==12.1.3.1
–nvidia-cuda-cupti-cu12==12.1.105
–nvidia-cuda-nvrtc-cu12==12.1.105
–nvidia-cuda-runtime-cu12==12.1.105
–nvidia-cudnn-cu12==8.9.2.26
–nvidia-cufft-cu12==11.0.2.54
–nvidia-curand-cu12==10.3.2.106
–nvidia-cusolver-cu12==11.4.5.107
–nvidia-cusparse-cu12==12.1.0.106
–nvidia-nccl-cu12==2.18.1
–nvidia-nvjitlink-cu12==12.3.52
–nvidia-nvtx-cu12==12.1.105

Really appreciate any help regarding this!

2

This forum is for the discussion of technical issues. For legal questions about the of software components associated with CUDA you would probably want to start by consulting the relevant End User License Agreement (EULA) or have your organization’s legal department do so.

Section 2.5 of the End User License Agreement specifies how to contact NVIDIA for questions regarding the EULA:

https://docs.nvidia.com/cuda/eula/index.html#licensing

Section 2.6 of the End User License Agreement lists the re-distributable components (“The following CUDA Toolkit files may be distributed with Licensee Applications developed by you …”):

https://docs.nvidia.com/cuda/eula/index.html#attachment-a

1 Like