# CUDA on slices

**URL:** https://discourse.julialang.org/t/cuda-on-slices/86782
**Category:** General Usage
**Tags:** cuda, broadcasting, cuarrays, tensors, mapslices
**Created:** [September 5, 2022, 5:35am UTC](https://discourse.julialang.org/t/cuda-on-slices/86782 "2022-09-05T05:35:28Z")
**Posts on this page:** 1
**Showing post:** 5

<div class="post-metadata">

### Author: ![Lincoln\_Hannah](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/lincoln_hannah/32/19198_2.png) [@Lincoln\_Hannah](https://discourse.julialang.org/u/Lincoln_Hannah)
#### Post date: [September 6, 2022, 2:18am UTC](https://discourse.julialang.org/t/cuda-on-slices/86782/5 "2022-09-06T02:18:08Z")

</div>

Thanks Jaratur.  
I tried a few einsum packages. When they do work they tend to be slower than `mapslices` on a CPU.

I’ve reposted here [Mapslices very slow](https://discourse.julialang.org/t/mapslices-very-slow/86826)

I’d rather not write a custom Kernel. See if there’s any alternatives.

---

_[View the full topic](https://discourse.julialang.org/t/cuda-on-slices/86782)._