Skip to content

Getting Started

Run GitHub Actions jobs on Avrea runners by changing runs-on in your workflow. Caching, observability, and infrastructure are handled automatically.

To have Avrea schedule and run the whole workflow, use Avrea CI. Start with the migration guide to copy workflows from .github/workflows/ to .avrea/workflows/ and review compatibility.

  1. Install the Avrea GitHub App

    Sign in at console.avrea.com, create or select an Avrea organization, then say where your repositories live:

    • GitHub organization installs the Organization App. Avrea registers runners in a runner group of your organization and does not ask for repository Administration access. See Organization runners.
    • Personal GitHub account installs the Repository App, which registers runners on each repository.

    GitHub then asks which account or organization to install on, and whether to grant access to all repositories or only a selection. Pick all your repos or just one to try it out; you can always add more later. See GitHub App permissions for what each App requests.

  2. Duplicate your workflow

    Create a copy of your existing workflow file and update the runner:

    workflow.yml
    jobs:
    build:
    runs-on: ubuntu-latest
    runs-on: avrea-ubuntu-latest

    Keep your original workflow unchanged so you can compare results side by side. avrea-ubuntu-latest matches GitHub's default (2 vCPUs, 8 GB RAM). For larger VM sizes, see Runners & Pricing.

    If you lint your workflow files with actionlint, it also needs to learn the new label. See Linting your workflows.

  3. Trigger both workflows

    Run each workflow and compare Avrea with your existing setup. Run it twice: the first run shows the hardware speedup and the second run is where the cache kicks in for a much larger improvement.

  • Avrea CI: run workflows directly on Avrea with GitHub Actions YAML.
  • Cache: transparent acceleration for GitHub Actions cache, build caches, and package managers.
  • Debugging & SSH: SSH into a running job from the console or the CLI.
  • Observability: step-level logs, run history, and failure/flake analytics.

Reach out at support@avrea.com for onboarding questions, technical issues, or product feedback. Or join the Avrea Slack community.