# JuliaGPU CI moving to Buildkite

**URL:** <https://discourse.julialang.org/t/juliagpu-ci-moving-to-buildkite/49925>\
**Category:** GPU\
**Tags:** ci\
**Created:** [November 10, 2020, 5:03pm UTC](https://discourse.julialang.org/t/juliagpu-ci-moving-to-buildkite/49925 "2020-11-10T17:03:40Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![maleadt](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/maleadt/32/10097_2.png) [@maleadt](https://discourse.julialang.org/u/maleadt)\
**Post date:** [November 10, 2020, 5:03pm UTC](https://discourse.julialang.org/t/juliagpu-ci-moving-to-buildkite/49925/1 "2020-11-10T17:03:40Z")

</div>

Hi all,

Because of GitLab limitations (no CI for external PRs, needs a repository mirror that can lose its sync, no real Julia support, only reuse of code through `include`, etc), Julia’s community GPU CI is moving to [Buildkite](https://buildkite.com/). Instructions on how to set-up your repository are available at [JuliaGPU/buildkite](https://github.com/JuliaGPU/buildkite). For an example of a full-featured pipeline, have a look at the [CUDA.jl pipeline.yml](https://github.com/JuliaGPU/CUDA.jl/blob/master/.buildkite/pipeline.yml) and [its execution](https://buildkite.com/julialang/cuda-dot-jl/builds/119); very similar to the old [GitLab pipeline](https://github.com/JuliaGPU/CUDA.jl/blob/v2.1.0/.gitlab-ci.yml), so migration shouldn’t be hard.

For now, I’ll be keeping a GitLab runner active, but with a very limited pool of runners. It’s thus recommended to switch your CI if you want fast jobs 🙂