Queen MQ is built at Smartness and developed in the open. The broker, the clients, the dashboard, the Laravel driver and supervisor, the benchmarks and these docs live in one repository, github.com/queen-mq/queen, and every change to them is a commit there.
Licence
Queen MQ is open source under the Apache License 2.0. That covers the broker and every client in the repository. Contributions are accepted under the same licence.
Releases and packages
The broker ships as one container image, ghcr.io/queen-mq/queen, and as a binary you can build
from source (run a node). Each release is listed, with what changed, on the
changelog and on
GitHub Releases. The clients are published to the
package registry of their language; clients has the install line for each.
Version 2.0.0 (2026-10-03) replaced the PostgreSQL storage of 1.x with Queen’s own log, replicated with Raft. Security fixes go to 2.x only: 1.x is not supported, and a 1.x deployment has to upgrade.
Bugs, questions and features
Open an issue with one of the issue templates. For a bug, include the broker version, how you run it (Docker, Helm or the binary) and the client you use. For a code change, open an issue first; contributing has the build, the tests and the rules a change to the engine, the docs or a benchmark follows.
Security
Do not report a vulnerability in a public issue. Report it privately through GitHub private vulnerability reporting, with the affected component and version, what the issue allows, and how to reproduce it if you can. We acknowledge the report, keep you informed while we fix it, and credit you in the release notes unless you prefer otherwise. How a deployment is secured is on security.
Hosted Queen
Queen Cloud runs Queen for you, and has a free tier.
The evidence behind the claims
The pages that describe the running system name, at their foot, the source files that govern them and the tests that exercise what they show, and a check fails the docs build when such a page is older than the code it describes. The performance claims come with the machines, the settings and the commands that produced them (how we measure), including the runs where another system does better. Correctness is tested with Jepsen, with every campaign, the faults it injected and the bugs it found.