Stop vs destroy
This is the most important distinction in the product, and the one that most often produces an unexpected bill. Read it once, properly.
The short version
| Stop | Destroy | |
|---|---|---|
| Container | shut down | removed |
| Your disk and its contents | kept | deleted, permanently |
| The GPU slot | held for you, unless the platform releases it (below) | released to the marketplace |
| GPU charges | stop | stop |
| Storage charges | continue | stop |
| Can you resume it? | yes, in place | no |
Stopping is a pause that still costs money. Destroying is final and costs nothing further.
Why stopping still bills you
When you stop an instance, your disk stays on that physical machine, and that machine cannot give the space away to anyone else. You are charged for that reservation at the storage rate — quoted per GB per hour — for as long as the instance exists in a stopped state.
The GPU itself is no longer charged, and keeping it reserved costs nothing extra: a stopped instance bills storage only. You are paying to keep your data and your claim on that slice. There is no queue to hold a place in — while your stop holds the GPUs, nobody else can rent them.
This means a large disk you forgot about is a slow, quiet leak. A 500 GB volume left stopped for a month costs real money even though nothing ran.
If you are finished with the work, destroy it. Stop is for "I'll be back this afternoon", not "I'm done".
Why the GPU slot matters
Stopping holds your slot on that specific machine. When you start again, you land back on the same hardware with the same disk — no re-download, no re-setup, and no risk that the offer was rented out from under you while you were away. The one exception is a release by the platform, after your balance reached zero or a payment dispute froze payments: see below.
Destroying releases the slot immediately. The offer goes back on the marketplace, and someone else may take it within seconds. There is no way to reclaim it.
When the platform releases the GPUs
Two things end the reservation without destroying anything, where the machine supports it:
- Your balance reaches zero. Running instances are stopped, and every stopped instance in the organization — including one you stopped yourself — has its GPUs released back to the machine.
- A payment dispute freezes the organization's payments. Starting is refused until the dispute closes, so holding GPUs for you would buy nothing: the GPUs of stopped instances are released, even with a positive balance, and a Stop you press while payments are frozen releases them straight away. The console says so before you confirm. Running instances keep running.
A release keeps the container and the disk on the machine, and the instance still bills storage only. What changes is Start. It has to re-claim the same GPUs on the same machine, because the container was created for them. If another renter is using any of them, or one has been taken out of service, the start is refused with 409 CAPACITY_UNAVAILABLE and the instance stays stopped with its disk. Try again later, or copy your data off and destroy it.
Where a machine cannot release GPUs, or the instance runs a vm template, they stay reserved exactly as after your own stop, and you still pay storage only.
The instance page shows GPUs released once they are back with the machine, and the API's gpus_released and gpu_release fields say the same. See Instances.
Which should you use?
Stop when all of these are true:
- You'll return within hours or a day or two.
- There's state on the disk you'd rather not rebuild — a dataset, a conda environment, checkpoints.
- The disk is small enough that the storage charge is cheaper than the time it would take to rebuild.
Destroy when any of these are true:
- The work is finished.
- Everything valuable is already pushed somewhere else — object storage, a git remote, a model registry.
- The disk is large, and re-downloading is cheaper than parking it.
Before you destroy
Destruction deletes the disk. There is no snapshot, no trash, and no undo. Copy anything you care about off the instance first:
# from your own machine — port, user and host are the ones in the
# connect command on the instance's Connect tab
scp -P <port> -r <user>@<host>:/workspace/checkpoints ./checkpoints
Or push it to wherever it belongs — S3, a Hugging Face repo, a git remote.
What the system does on its own
Two automatic actions apply the same distinction, to stop a forgotten instance from draining an account to nothing:
- When your balance reaches zero, running instances are stopped, and stopped instances have their GPUs released where the machine supports it. Your data survives; storage charges continue against a balance that is already empty.
- If the balance then stays at or below zero for 24 hours and is at or below −$5, the stopped instances are destroyed. At that point the data is gone.
Top up before then. See How billing works.