Manage Runners
New: Autoscaling on Pro

GitLab and GitHub runners on your own Hetzner or DigitalOcean

Runners are servers in your Hetzner project or DigitalOcean account, registered with GitLab.com, self-managed GitLab or GitHub.com. Create, duplicate, pause and autoscale them from the dashboard, CLI or AI assistant. Track CPU, RAM, swap, disk and running jobs in the dashboard and, optionally, your own OpenTelemetry backend.

From $5.99/month€5.90/month excl. VAT for unlimited runners, plus the servers, which Hetzner or DigitalOcean bills to you directly.

A runner card in the Manage Runners dashboard: an active GitLab runner with charts of its CPU, RAM, swap, running jobs and disk usage over the last 24 hours

How it fits together

Manage Runners sets up and looks after GitLab and GitHub runners in your Hetzner Cloud project or DigitalOcean account. Your CI jobs run between your GitLab or GitHub and your own server.

Yours

Your GitLab or GitHub

GitLab.com, self-managed GitLab or GitHub.com. Holds your code and pipelines and hands jobs to the runner.

Registers as a runner and fetches jobs

Yours

Server in your cloud account

Debian 13 with Docker and GitLab Runner or the GitHub Actions runner, on Hetzner Cloud or DigitalOcean. Runs your CI jobs. Your provider bills you for it directly.

Creates, pauses and resumes it with your Hetzner or DigitalOcean API token

Ours

Manage Runners

Dashboard, CLI and MCP server. Configures the server, schedules it and shows its status.

One runner is one server running GitLab Runner or the GitHub Actions runner

Executor
GitLab runners use Docker, Docker-in-Docker or Shell; with Docker, each job runs in its own container. GitHub jobs run on the server, with Docker available.
Concurrency
How many jobs it runs at once: GitLab Runner's concurrent setting, or one GitHub runner instance per job. The dashboard sets it to twice the server's vCPU count; you can change it. GitHub runners have at most 32.
Capacity
On Pro and Enterprise, autoscaling adds burst runners to a busy runner within minutes and drains them when the load falls. Hetzner bills them by the started hour, DigitalOcean per second; autoscaling never stops the runner itself. You can also duplicate runners, and pause or schedule them to stop paying for idle hours.
  • Your jobs never pass through us

    GitLab Runner or the GitHub Actions runner on your server fetches jobs from your GitLab or GitHub and runs them there. Code, job logs, CI/CD variables and secrets stay between your GitLab or GitHub and your server.

  • Only what your API token allows

    Manage Runners creates, pauses, resumes and deletes runner servers and their IP addresses through the Hetzner Cloud or DigitalOcean API. It creates no firewalls, networks, volumes or SSH keys.

  • No SSH access for us

    Only the SSH keys you select from your Hetzner project or DigitalOcean account are installed. After setup, the server contacts Manage Runners only to report metrics, if you turn them on.

What you can do with it

Setup

Create a runner from one form

Enter a GitLab runner token, or a GitHub repository or organization with a fine-grained token, and a Hetzner or DigitalOcean API token, pick a server type and location, and the server installs Docker and the runner and registers within a few minutes. No SSH session or config files.

Capacity

Duplicate for more capacity

Copy a runner with its configuration and its GitLab or GitHub token. The copy gets its own server and IP addresses and takes jobs alongside the original. On Pro and Enterprise, autoscaling adds burst runners to a busy runner for you.

Cost

Pause and schedule

Pausing deletes the server and keeps the runner's settings, so your provider stops billing for the server, though kept IP addresses may still cost something: on DigitalOcean USD 5 a month for the reserved IPv4. Running jobs can finish first, for up to an hour. Schedules pause and resume runners on a weekly timetable; set them with the CLI, the API or your AI assistant, not yet in the dashboard. Schedules need the Pro or Enterprise plan.

Network

Fixed IP addresses

Each runner keeps its IPv4 and IPv6 addresses across pause and resume, so allow lists and firewall rules on your GitLab or deployment targets keep working.

Maintenance

Automatic updates

Servers install Debian, Docker and GitLab Runner package updates on their own and reboot when an update needs it; the GitHub Actions runner updates itself. Docker stays on major version 29. A resumed runner starts on a freshly installed server.

Visibility

Metrics, in the dashboard or yours

See each runner's CPU, RAM, swap and disk usage and running jobs in the dashboard, or have the runner push them over OpenTelemetry to your own monitoring, such as Grafana Cloud or OpenObserve.

3 steps to your first runner

If you already have a Hetzner Cloud or DigitalOcean account and a GitLab or GitHub account, the quick start takes you from sign-up to your first CI job in about 15 minutes, including creating the two tokens.

01

Create two tokens

Create a runner in GitLab and copy its runner authentication token, or a fine-grained GitHub token for your repository or organization. Then create an API token with Read & Write access in your Hetzner Cloud project, or with the listed custom scopes in your DigitalOcean account.

02

Configure your runner

Pick GitLab or GitHub, a provider, server type, location and concurrency, and for GitLab an executor. Optionally select SSH keys from your Hetzner project or DigitalOcean account, and add labels, which GitHub registers and a Hetzner firewall can apply to.

03

Run your jobs

The server installs Docker and GitLab Runner or the GitHub Actions runner and registers within a few minutes. GitLab jobs with the runner's tags, or untagged jobs if you allow them, and GitHub jobs whose runs-on matches its labels then run on it.

Manage runners from your AI assistant

Manage Runners is a remote MCP server for Claude, Claude Code and other clients that support Streamable HTTP and MCP authorization. Sign in with your account and choose what your assistant may do.

  • Reads your runners, metrics, schedules, scaling policies and server prices.
  • Previews creating, resuming, duplicating or replacing a server for you to confirm.
  • Secrets never pass through the AI: you enter tokens in the dashboard.

Connector URL

https://api.managerunners.com/mcp

Claude Code

claude mcp add --transport http manage-runners https://api.managerunners.com/mcp

Frequently Asked Questions

What does it cost in total?

Two bills. Manage Runners charges your organization's plan: Basic is $5.99/month€5.90/month excl. VAT for unlimited runners, and Pro at $19.99/month€19.90/month excl. VAT adds schedules and autoscaling. Your cloud provider bills you directly for each runner's server: Hetzner by the hour up to a monthly price (see Hetzner's pricing), DigitalOcean per second up to a monthly price (see DigitalOcean's pricing). A paused runner has no server, but Hetzner may still charge for its kept IP addresses, and DigitalOcean charges USD 5 a month for its kept reserved IPv4. See our pricing.

Why my own servers instead of GitLab's or GitHub's hosted runners?

GitLab's hosted runners use compute minutes (see GitLab's pricing), and GitHub's hosted runners are priced per minute (see GitHub's runner pricing). Your own server costs the same however many jobs it runs: Hetzner bills it by the hour, DigitalOcean per second, each up to a monthly price. Which is cheaper depends on how much CI you run and how many hours your runners are active; for GitLab, see the GitLab CI cost comparison. On your own server, Docker images, and on GitLab runners the local CI cache, can be reused between jobs, subject to automatic cleanup, and are lost when the server is recreated. You also choose the server type, location and firewall rules.

How much can I save compared with hosted runners?

It depends on how many minutes you use. Beyond what your plan's included compute minutes cover, GitLab.com bills $10.00 per 1,000 job minutes on the small Linux runner and up to $120.00 on 2xlarge. A Hetzner server costs the same however many jobs it runs. For example, at prices as of 19 October 2026, the cheapest server matching the small runner, CX33 in Germany/Finland, costs $10.59 a month including IPv4, plus $5.99 for the Basic plan. Always on, that is cheaper than hosted minutes from about 1,658 small-runner job minutes a month beyond those the quota covers; below that, hosted runners cost less. Compare your own GitLab usage with the calculator. The calculator covers GitLab; for GitHub Actions, compare GitHub's per-minute rates with the server prices the dashboard shows in the same way.

Does it autoscale?

Yes, for GitLab and GitHub runners on the Pro and Enterprise plans. Autoscaling adds burst runners, extra Hetzner servers or DigitalOcean droplets, when few of a runner's job slots are free, and drains them when the load falls. It reacts within a few minutes, not seconds, and never stops the runner you set it up on. Hetzner bills each burst runner by the started hour, DigitalOcean per second, including time spent draining. See Autoscaling. To stop paying for idle hours, pause runners by hand or with a schedule (Pro and Enterprise plans).

How many jobs can a runner run at once?

As many as its concurrency. On a GitLab runner, that is GitLab Runner's global concurrent setting. On a GitHub runner, it is the number of GitHub Actions runner instances on the server, 1 to 32, each running one job at a time; see GitHub setup. Choosing a server type in the dashboard sets it to twice the server's vCPU count, at most 32 on GitHub; lower it for memory-hungry jobs. Changing it recreates the server. See Concurrency.

If I pause a runner, will it still cost me anything?

Pausing deletes the runner's server and keeps its configuration and its IPv4 and IPv6 addresses. You can let running jobs finish first: the runner takes no new jobs, and the server is deleted once they are done, at the latest after an hour, so your provider bills it until then. After that, the server costs nothing, but Hetzner may still charge for the kept IP addresses (see Hetzner's pricing), and DigitalOcean charges USD 5 a month for the kept reserved IPv4. Resuming creates a fresh server with the same addresses. While a GitHub runner is pausing, GitHub can still give it jobs that ask only for the default labels, so give your workflows one of its own labels.

What happens to my runners if I cancel my plan?

Cancelling doesn't delete your runners' servers: they stay in your Hetzner project or DigitalOcean account, and the provider keeps billing them. Burst runners added by autoscaling are the exception: once the plan ends, none are added, and existing ones start draining, on Hetzner in the last 15 minutes of a paid hour, and get up to one hour to finish running jobs before they are deleted. Enabled schedules keep pausing runners at their stop times, which deletes the server, but scheduled starts stop without an active plan. To keep a runner running on its own, delete its schedule before the plan ends; schedules can't be changed afterwards. To stop the provider's charges, delete the runners with the CLI or API, which deletes their servers and IP addresses.

What happens if Manage Runners is down?

Active runners keep running jobs: GitLab Runner or the GitHub Actions runner on your server takes jobs from your GitLab or GitHub, not from us. Until Manage Runners is back, the dashboard, CLI and MCP server are unavailable, schedules don't pause or resume runners, autoscaling adds and removes no burst runners, and the dashboard records no metrics. Metrics your runners send to your own OpenTelemetry endpoint keep going.

Can I put my runners behind a firewall?

Yes. Manage Runners creates no firewalls. GitLab Runner and the GitHub Actions runner only make outgoing connections, so you can block all incoming traffic, apart from SSH if you log in. Outgoing, the server needs your GitLab, or the GitHub addresses listed in GitHub setup; the Debian and Docker package repositories, and on GitLab runners GitLab Runner's, for setup and automatic updates; Manage Runners during setup and for metrics; your OpenTelemetry endpoint if you send metrics there; and whatever your jobs reach. On Hetzner, the labels you give a runner become server labels, so a firewall that applies to a label selector covers your runners; see Labels and firewalls. DigitalOcean droplets get no labels, so a DigitalOcean firewall cannot select runners by label; see DigitalOcean setup.

Will you SSH into my servers?

No. Only the SSH keys you select from your Hetzner project or DigitalOcean account are added, for root, and password login is disabled. Manage Runners adds no keys of its own.

What does my runner send to Manage Runners?

During setup, the server reports when it starts configuring and when GitLab Runner or the GitHub Actions runner is registered, so the dashboard can show its state. After that it contacts Manage Runners only for metrics, if you turn them on: about once a minute its CPU, RAM, swap and disk usage and number of running jobs for the dashboard, or, if it only sends metrics to your own endpoint, the result of its last export. Your jobs, code, logs, CI/CD variables and secrets are never sent.

Which server types and locations can I use?

On Hetzner, every Hetzner Cloud server type that Hetzner has not deprecated and offers in at least one location, with x86 (Intel or AMD) or ARM CPUs, in Falkenstein, Nuremberg, Helsinki, Singapore, Ashburn and Hillsboro. On DigitalOcean, the Basic, General Purpose and CPU-Optimized droplet sizes with at least 1 GB of memory, all x86, in Amsterdam, Frankfurt, London, New York, San Francisco, Atlanta, Toronto, Singapore, Bangalore and Sydney. Not every type is offered everywhere; the dashboard shows where each one is available and at what price.

How do I choose between Hetzner and DigitalOcean?

Both run GitLab and GitHub runners with pause and resume, fixed IP addresses, schedules, autoscaling and metrics, and you choose the provider per runner. The main differences: Hetzner bills by the started hour, DigitalOcean per second, so autoscaling removes DigitalOcean burst runners without waiting for the end of an hour. Hetzner offers x86 and ARM servers, DigitalOcean only x86 droplets. A Hetzner runner's labels become server labels that a Hetzner firewall can select; droplets get no labels. See Hetzner setup and DigitalOcean setup.

Which GitLab and GitHub instances does it work with?

GitLab.com and self-managed GitLab: create a runner in GitLab to get its runner authentication token (glrt-…) and enter your instance's URL. A self-managed instance must be reachable from the runner's IP addresses, which stay the same when you pause and resume, so you can allow them in your firewall. See GitLab setup. On GitHub, repository and organization runners on GitHub.com, not GitHub Enterprise Server: enter the repository or organization and a fine-grained personal access token that may manage its self-hosted runners. See GitHub setup.

Who is behind Manage Runners, and where is my data stored?

Manage Runners is run by Detech Digital Co., Ltd. in Hanoi, Vietnam (legal notice). According to the privacy policy, your data is hosted by Hetzner Online GmbH in data centers in Germany, administrators in Vietnam may access it for support and service delivery under Standard Contractual Clauses, and Cloudflare protects and delivers the website. Payments go through Dodo Payments. Your CI jobs, code and logs stay between your GitLab or GitHub and your server.

Run your CI on your own Hetzner or DigitalOcean servers

Plans start at $5.99/month€5.90/month excl. VAT for unlimited runners. Hetzner or DigitalOcean bills the servers to your own account.