aq.

PricingDocsGuidesCompare

Sign inStart free

PricingDocsGuidesCompareSign in

Claude Code · Codex · Cursor · Kimi · Grok

The multiplayer agentic
development environment
that runs on your infrastructure.

Engineers, PMs, and designers run coding agents together: shared live terminals, a code editor, and app previews, in any browser. Your own cloud, a dedicated AQ-managed VM, or a free sandbox to start.

Start freeTalk to sales

Work-email sign-in. Every request personally reviewed.

● Free sandbox, or your own cloudThe real CLIs, your accountsWeb + mobile, any OS

[An AQ workspace in the browser: the team sidebar with live p]

Live multiplayer

Your whole team, inside the session.

Teammates open your workspace and watch the same live terminal, not a replay. Hand someone the keyboard with one click: they type as you, you keep your own input, and you can revoke at any time. When an agent needs a decision, the person with the answer is already in the room.

[A teammate requests terminal control and the session owner g]

Parallel workspaces

Every task gets its own worktree.

Each workspace is an isolated git worktree, so parallel agents never step on each other. Dependencies install automatically, and one click rebases the branch onto main. Sessions run in persistent tmux on the VM: close the laptop, nothing stops.

[The team sidebar: five active workspaces with live presence,]

Live previews

Review the running app, not just the diff.

A workspace can run its dev server and stream the app right next to the agent. On the Team plan, share it with a link that opens without an account for viewing, pin comments on the page itself, and send a comment to the agent as its next prompt.

[A live app preview inside the workspace with annotate mode a]

Linear in, PR out

From issue to pull request without a handoff.

Label a Linear issue ai-task and a workspace appears in the team sidebar, with status syncing both ways as work moves. Agents commit, push, and open pull requests with each person's own GitHub auth, and every PR is tracked on its workspace.

[Workspace rows created from Linear issues, showing issue ide]

Any device, machines dedicated to you

A browser tab, backed by real infrastructure.

AQ runs where you are: macOS, Windows, Linux, and your phone. Agents sign in with each person's own accounts, and AQ never marks up model usage. Work runs on VMs from your own cloud, a dedicated AQ-managed machine, or a free personal sandbox, and sessions are exactly where you left them, from any device.

[The same Claude Code session open on a phone, with the mobil]

Up and running in three steps.

1

Connect an environment

Start free in a personal sandbox AQ creates for you, or subscribe to run in your own cloud with a dedicated, always-on VM for your team.

2

Sign in to your agents

Claude Code, Codex, Cursor Agent, and every other agent CLI, with the accounts you already have. AQ never marks up your model usage.

3

Build together, live

Every task gets its own workspace: shared live terminals, a code editor, and an app preview your whole team opens with a link.

Frequently asked questions

What is included in the Free plan?

AQ's Free plan has no seat charge and gives one person a private, AQ-managed sandbox with nothing to install. Model usage is separate: use your own agent subscriptions, or use AQ Models with AQ credits. The Team plan adds teammates and sharing, plus your choice of VMs in your own cloud or a dedicated, always-on VM managed by AQ. During early access, Team is $50 per user per month.

Do I need my own Claude or OpenAI subscription?

Only if you want to use Claude Code or Codex. Each person signs in to those CLIs with their own Claude or OpenAI account, and AQ does not proxy or mark up that usage. You can instead run OpenCode or Pi with AQ Models such as GLM 5.2. That inference goes through AQ's gateway and consumes AQ credits.

Where does my code run?

With the Free plan, your code runs on a private AQ-managed VM dedicated to you. With the Team plan, it runs either on VMs you connect from your own cloud account or on a dedicated AQ-managed VM. Your checkout, worktrees, shell sessions, and CLI credentials stay on that VM; AQ does not use shared multi-tenant execution infrastructure. Model inference is separate: prompts may send code context directly to the model provider or through AQ's gateway when you use AQ Models.

Is AQ an agentic development environment?

Yes, by anatomy: workspaces as the task surface, one isolated git worktree per task, the real agent CLIs as the runtime, a live terminal and editor for stepping in, and pull requests out the other side. The difference from desktop ADEs is the axis they mostly skip: AQ sessions are multiplayer, so your team can watch and steer the same live session, and everything runs on machines dedicated to you instead of one person's laptop.

What happens when I click “Start free”?

You accept the Terms of Service and continue with a work Google account. If you are not already approved, AQ adds you to the review queue. Requests are reviewed personally, usually within a day or two, and AQ emails you when your workspace is ready. Then sign in, connect an agent, and add your first project.

Bring your whole team into the workspace.

Engineers, PMs, and designers, building with agents in one place, live. Share a link and your team is instantly inside. All in your own cloud. Start free in a sandbox AQ creates for you, run it in your own cloud, or let AQ run a dedicated, always-on VM for your team at $50 per user per month.

Start freeTalk to sales