# Using Julia to install system wide CUDA stack?

**URL:** https://discourse.julialang.org/t/using-julia-to-install-system-wide-cuda-stack/78257
**Category:** General Usage
**Tags:** question
**Created:** [March 22, 2022, 8:18am UTC](https://discourse.julialang.org/t/using-julia-to-install-system-wide-cuda-stack/78257 "2022-03-22T08:18:09Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![jw3126](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jw3126/32/3086_2.png) [@jw3126](https://discourse.julialang.org/u/jw3126)
#### Post date: [March 22, 2022, 8:18am UTC](https://discourse.julialang.org/t/using-julia-to-install-system-wide-cuda-stack/78257/1 "2022-03-22T08:18:10Z")

</div>

It is amazing how a single `]add CUDA` automatically installs a private CUDA stack in a pain-free way. Sometimes however I need a CUDA installation for non-julia applications, like pytorch etc. Is there a way to get a “global” CUDA installation using julia? For instance, a script that copies the julia CUDA artifacts to `/usr/lib` or whatever is the appropriate library directory for the current platform?