Supported Platforms and System Requirements

Antora runs on Linux, macOS, and Windows. Thanks to this broad platform support, Antora is also used on many cloud platforms and continuous integration and delivery servers.


Our general policy is to align Antora’s support of a platform version with the upstream project’s lifecycle schedule for that version. When a platform version reaches the end of active support by its maintainers or its end of life (EOL)—​which ever comes first—​Antora no longer supports it.

Supported Platform Supported Version(s)

Alpine Linux

≥ 3.13


≥ 10.12


≥ 35


18.04 LTS
20.04 LTS


Catalina (10.15)
Big Sur (11.6)
Monterey (12.4)


Windows 10
Windows 11
Windows Server 2016
Windows Server 2019


Latest stable version


Latest stable version

MS Edge

Latest stable version


16 LTS

About the Antora Test Suite

At last count, Antora had over 2008 tests that cover 99% of the lines. You could say we’re just a teeny-tiny bit obsessed with writing high quality tests and improving test coverage.

Cloud platforms

Antora should operate as designed on common cloud platforms. How you provision your cloud instance depends on your workload requirements and remote git repository sizes. If you’re running Antora on a cloud platform, we would love to hear about your experience in Antora’s community chat.

Hardware recommendations

Your hardware requirements will depend primarily on the size of the remote git repositories your Antora pipeline fetches content from when it is generating your site.

The first time Antora runs, it fetches and clones all of the remote git repositories specified in a playbook. The speed of this operation is dictated by the size of the remote repositories, environment input/output parameters, and network performance. After the initial run, Antora caches the repositories locally. On subsequent runs, Antora only reconnects to the remote repositories if the fetch option is enabled or the cache folder is removed.


Memory requirements depend on the size of your Git repositories. 3GB should provide sufficient headroom.


Maximum throughput and minimum latency always make things nicer but aren’t required.

Learn more