Auto-pause
Walk away from a GPU and it warns you, saves your setup, and releases the machine — so an instance you forgot about stops billing instead of running all weekend.
Provider coverage
Works the same way on every provider — auto-pause saves your setup, then releases the machine using the same mechanism manual Pause does, not a provider-specific stop API. Idle detection reads the same utilization data as Live GPU Metrics, so a GPU profile that can't report utilization (for example, Vultr's vGPU-based A16 line) shows as "we can't read utilization on this box" instead of being guessed at.
Off until you turn it on
Nothing you deploy is watched unless you ask for it. In the console deploy sheet, switch on Auto-pause when idle and the deployment is watched from the moment it comes up. You can also turn it on later from that deployment's Idle tab. Leave it off and nothing is watched or paused.
It is per deployment, never account-wide, and it is never switched on retroactively for boxes you already have running. A deploy made through the API or the aq CLI has it off unless you set the policy explicitly.
The toggle works the same way on every provider — it saves your setup, then releases the machine using the same create-and-terminate mechanism manual Pause uses, not a provider-specific stop API, so there's no per-provider list to check before you rely on it. Resume later brings that setup back on a fresh machine, possibly on a different provider — the same round trip manual Pause and Resume use.
Left at their defaults, the thresholds are 5% GPU utilization, a warning at 30 minutes idle, and auto-pause at 60 minutes.
Warn, then auto-pause
Set a GPU utilization threshold that counts as idle, how many minutes of that before you're warned, and how many minutes before it acts. Utilization at or under the threshold for the warn period sends an email — the box is still running and still billing at that point. Staying idle past the act threshold saves your setup first, then releases the machine — the same idle-triggered auto-stop behavior the product now calls auto-pause.
The warn threshold always has to sit below the act threshold; the console won't save a configuration where the warning would fire after the box was already paused.
Four states, and unknown is a real one
A deployment with auto-pause configured shows one of: in use, idle, idle past the auto-pause threshold, or we can't read utilization on this box. That last one is not a rounding of "probably idle" — it's an explicit unknown, shown whenever utilization or on-box presence can't be read, and it never triggers an auto-pause.
Presence is read from the on-box agent, which sees SSH logins and browser-terminal sessions. The system doesn't infer that nobody is on a box from silence — an unreadable signal stays unknown rather than being treated as idle.
FAQ
Is auto-pause on by default?
No — it is off unless you turn it on, on every surface. In the console, switch on "Auto-pause when idle" in the deploy sheet before deploying, or turn it on later from the deployment's Idle tab; through the API or the aq CLI, set the idle policy explicitly. Once on, it works the same way on every provider: it saves your setup then releases the machine, the same mechanism manual Pause uses everywhere. It is per deployment rather than account-wide.
What happens right before it pauses the box?
It saves your setup to storage first, the same save mechanism used for a manual Pause, then releases the machine. It's a file save, not a resume point for whatever was running in memory.
What does "we can't read utilization" mean?
It means the platform genuinely can't tell whether the box is idle or in use — the utilization reading or the on-box presence signal is unavailable. It's shown as its own state and never treated as idle, so it can't trigger an auto-pause on a guess.
Can I change the thresholds after deploying?
Yes. Open the deployment's Idle tab to turn auto-pause on or off, or change the utilization threshold, warn minutes, and act minutes at any time.
Does it work the same on every provider?
Yes. Saving and releasing the machine uses the same create-and-terminate mechanism manual Pause uses on every provider, not a provider-specific stop API, so there's no per-provider list to check. Idle detection itself reads the same utilization data as Live GPU Metrics everywhere, so a GPU whose virtualization profile can't report utilization shows as unknown rather than being marked idle.
