Skip to content

GitLab-hosted runners

DETAILS: Tier: Free, Premium, Ultimate Offering:, GitLab Dedicated

You can run your CI/CD jobs on and GitLab Dedicated using GitLab-hosted runners to seamlessly build, test and deploy your application on different environments.

Hosted runners for

DETAILS: Offering:

These runners fully integrated with and are enabled by default for all projects, with no configuration required. Your jobs can run on:

How hosted runners for work

When you use hosted runners:

  • Each of your jobs runs in a newly provisioned VM, which is dedicated to the specific job.
  • The virtual machine where your job runs has sudo access with no password.
  • The storage is shared by the operating system, the image with pre-installed software, and a copy of your cloned repository. This means that the available free disk space for your jobs to use is reduced.
  • Untagged jobs run on the small Linux x86-64 runner.

NOTE: Jobs handled by hosted runners on time out after 3 hours, regardless of the timeout configured in a project.

Security of hosted runners for

The following section provides an overview of the additional built-in layers that harden the security of the GitLab Runner build environment.

Hosted runners for are configured as such:

  • Firewall rules only allow outbound communication from the ephemeral VM to the public internet.
  • Inbound communication from the public internet to the ephemeral VM is not allowed.
  • Firewall rules do not permit communication between VMs.
  • The only internal communication allowed to the ephemeral VMs is from the runner manager.
  • VM isolation between job executions.
  • Ephemeral runner VMs will only serve a single job and will be deleted right after the job execution.

Architecture diagram of hosted runners for

The following graphic shows the architecture diagram of hosted runners for

Hosted runners for architecture

For more information on how runners are authenticating and executing the job payload, see Runner Execution Flow.

Job isolation of hosted runners for

In addition to isolating runners on the network, each ephemeral runner VM only serves a single job and is deleted straight after the job execution. In the following example, three jobs are executed in a project's pipeline. Each of these jobs runs in a dedicated ephemeral VM.

Job isolation

The build job ran on runner-ns46nmmj-project-43717858, test job on f131a6a2runner-new2m-od-project-43717858 and deploy job on runner-tmand5m-project-43717858.

GitLab sends the command to remove the ephemeral runner VM to the Google Compute API immediately after the CI job completes. The Google Compute Engine hypervisor takes over the task of securely deleting the virtual machine and associated data.

For more information about the security of hosted runners for, see Google Cloud Infrastructure Security Design Overview whitepaper, GitLab Trust Center, or GitLab Security Compliance Controls.

Caching on hosted runners for

The hosted runners share a distributed cache stored in a Google Cloud Storage (GCS) bucket. Cache contents not updated in the last 14 days are automatically removed, based on the object lifecycle management policy. The maximum size of an uploaded cache artifact can be 5 GB after the cache becomes a compressed archive.

For more information about how caching works, see Architecture diagram of hosted runners for, and Caching in GitLab CI/CD.

Pricing of hosted runners for

Jobs that run on hosted runners for consume compute minutes allocated to your namespace. The number of minutes you can use on these runners depends on the included compute minutes in your subscription plan or additionally purchased compute minutes.

For more information about the cost factor applied to the machine type based on size, see cost factor.

SLO & Release cycle for hosted runners for

Our SLO objective is to make 90% of CI/CD jobs start executing in 120 seconds or less. The error rate should be less than 0.5%.

We aim to update to the latest version of GitLab Runner within a week of its release. You can find all GitLab Runner breaking changes under Deprecations and removals.

Hosted runners for GitLab community contributions

DETAILS: Offering:

If you want to contribute to GitLab, jobs will be picked up by the fleet of runners, dedicated for GitLab projects and related community forks.

These runners are backed by the same machine type as our small Linux x86-64 runners. Unlike hosted runners for, hosted runners for GitLab community contributions are re-used up to 40 times.

As we want to encourage people to contribute, these runners are free of charge.

Hosted runners for GitLab Dedicated

DETAILS: Offering: GitLab Dedicated Status: Beta

These runners are created on demand for GitLab Dedicated customers and are fully integrated with your GitLab Dedicated instance. For more information, see hosted runners for GitLab Dedicated.

Supported image lifecycle

Hosted runners on macOS and Windows can only run jobs on supported images. You cannot bring your own image. Supported images have the following lifecycle:


New images are released as beta. This allows us to gather feedback and address potential issues before general availablility (GA). Any jobs running on beta images are not covered by the service-level agreement. If you use beta images, you can provide feedback by creating an issue.

General availability (GA)

A image becomes generally available after the image completes the beta phase and is considered stable. To become GA, the image must fulfill the following requirements:

  • Successful completion of a beta phase by resolving all reported significant bugs
  • Compatibility of installed software with the underlying OS

Jobs that run on GA images are covered by the defined service-level agreement.


A maximum of two generally available (GA) images are supported at a time. After a new GA image is released, the oldest GA image becomes deprecated. A deprecated image is no longer updated and is deleted after 3 months in accordance with the deprecation guidelines.