Skip to content

docs: Document Pod experience updates: Deploy When Available, savings plan discontinuation, Recommended/NVLink, Pod inspector #662

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
promptless wants to merge 1 commit into from
Closed

docs: Document Pod experience updates: Deploy When Available, savings plan discontinuation, Recommended/NVLink, Pod inspector #662

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions get-started.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,10 @@ Now that you've created your account, you're ready to deploy your first Pod:
If you haven't set up payments yet, you'll be prompted to add a payment method and purchase credits for your account.
</Note>

<Tip>
If your selected GPU is out of capacity, click **Deploy When Available** instead of **Deploy On-Demand**. Runpod queues your Pod and deploys it automatically once the GPU frees up, then notifies you by email and in your console inbox. See [Choose a Pod](/pods/choose-a-pod#deploy-when-available) for details.
</Tip>

</Tab>

<Tab title="Terminal">
Expand Down
26 changes: 25 additions & 1 deletion pods/choose-a-pod.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,7 +61,31 @@ For data-intensive workloads, ensure sufficient volume disk or network volume ca

1. **Right-size your resources**: Start with the minimum viable configuration, then scale up based on actual usage. Development and testing often need less power than production.

2. **Consider savings plans**: For extended usage, Runpod's [savings plans](/pods/pricing#savings-plans) reduce costs for committed usage.
2. **Use Deploy When Available for scarce GPUs**: When a high-demand GPU is out of capacity, [Deploy When Available](#deploy-when-available) lets you queue your Pod and deploy automatically once it frees up, instead of paying for a larger or pricier GPU to start sooner.

## Deployment options

When you deploy a Pod, the deploy screen surfaces a few features that help you find available capacity and choose the best GPU for your workload.

### Recommended tab

@promptless promptless Bot Jun 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Described the Recommended tab behavior from CON-34: it shows the template's recommended GPUs first, then adds up to three additional available GPUs ranked by VRAM (stockStatus not null) that aren't already recommended.

Source: https://linear.app/runpod/issue/CON-34/show-high-availability-options-in-recommended


The **Recommended** tab highlights the GPUs best suited to your selected template. It lists the template's recommended GPUs first, then adds up to three more GPUs that are currently available, ranked by VRAM. This lets you compare strong recommendations against options that are in stock and ready to deploy right away.

### NVLink availability

@promptless promptless Bot Jun 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sourced the NVLink availability banner for multi-GPU deploys (GPU count > 1) on H100/H200/B200, and that only SXM and NVL cards support NVLink, from CON-9 acceptance criteria and Greg Wester's note that "NVL and SXM are advertisements for nvlink."

Source: https://linear.app/runpod/issue/CON-9/show-nvlink-availability-for-multi-gpu-pod-deploys


When you deploy a multi-GPU Pod (a GPU count greater than 1) on an H100, H200, or B200, a banner tells you whether the GPU supports NVLink. Only SXM and NVL cards provide NVLink for high-bandwidth GPU-to-GPU communication, which benefits distributed training and other workloads that exchange large amounts of data between GPUs.

### Deploy when available

@promptless promptless Bot Jun 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Documented the Deploy when available window options from CON-152: clicking the Now or In 24 hours block opens a calendar selector with hour/minute precision, and 24 hours is the default. CON-132 establishes the feature exists for out-of-capacity GPUs with email + inbox notifications.

Source: https://linear.app/runpod/issue/CON-152/update-deploy-when-available-from-and-to-ux


When your chosen GPU is out of capacity, you don't have to deploy immediately or keep retrying. Select **Deploy When Available** to queue your Pod, and Runpod deploys it automatically as soon as the GPU frees up.

You set a window for when Runpod should attempt the deployment:

@promptless promptless Bot Jun 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sourced the email + console inbox notification "when your Pod deploys or finishes migrating" from CON-10 (notify when a Pod becomes available during migrations) combined with CON-132's email + inbox notification support.

Source: https://linear.app/runpod/issue/CON-10/notify-when-a-pod-becomes-available-during-migrations


- **Now**: Deploy as soon as capacity becomes available.
- **In 24 hours** (default): Keep trying for the next 24 hours.
- **Custom**: Use the calendar picker to set a specific From and To time, with hour and minute precision.

Runpod sends an email and a console inbox notification when your Pod deploys or finishes migrating, so you know when it's ready without watching the console.

## Secure Cloud vs Community Cloud

Expand Down
6 changes: 6 additions & 0 deletions pods/manage-pods.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -249,6 +249,12 @@ Pods provide two log types:

Access logs from the [Pods page](https://www.console.runpod.io/pods) by expanding your Pod and clicking **Logs**.

## View template details

@promptless promptless Bot Jun 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added the View template details section from CON-108: in the Pod inspector, "Template readme" was renamed to "Template details" and a "Report problem with template" button was added that opens an open-ended feedback modal.

Source: https://linear.app/runpod/issue/CON-108/add-template-feedback-inside-the-pod-details


Expand a Pod and open the **Template details** section to review the template your Pod was deployed from, including its readme and configuration.

If the template isn't working as expected, click **Report problem with template** to open a feedback form and describe the issue. Your feedback goes directly to the Runpod team to help improve the template.

## Troubleshooting

| Issue | Solution |
Expand Down
2 changes: 1 addition & 1 deletion pods/overview.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ Runpod offers two cloud options:

## Pricing

Pods are billed by the minute with no fees for ingress/egress. Runpod also offers long-term [savings plans](/pods/pricing#savings-plans) for extended usage patterns. See [Pod pricing](/pods/pricing) for details.
Pods are billed by the minute with no fees for ingress/egress. See [Pod pricing](/pods/pricing) for details.

## Limitations

Expand Down
6 changes: 3 additions & 3 deletions pods/pricing.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,12 +33,12 @@ You must have at least one hour's worth of credits for your selected configurati

### Savings plans

Commit to a 3-month or 6-month term upfront for significant discounts on compute costs. When you stop a Pod, the savings plan automatically applies to your next deployment of the same GPU type.

<Warning>
Savings plans only cover GPU compute costs—[storage costs](/pods/storage/types) are billed at standard rates. Storage charges continue to accrue on stopped Pods. If your balance reaches \$0, your Pods stop: those with a network volume are preserved, while those without one are terminated and their data cannot be recovered. Plans are non-refundable and have fixed expiration dates.
Savings plans have been discontinued. You can no longer create new savings plans, but existing plans are unaffected and remain active until their commitment period expires.

@promptless promptless Bot Jun 17, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Rewrote the Savings plans section as a discontinuation notice per CON-171: savings plan creation is being hidden/disabled while existing plans remain unaffected and run to expiration (~300 affected users notified by email).

Source: https://linear.app/runpod/issue/CON-171/discontinue-savings-plans

</Warning>

Savings plans let you commit to a 3-month or 6-month term upfront in exchange for discounts on compute costs. If you have an active plan, it continues to apply automatically: when you stop a Pod, the plan applies to your next deployment of the same GPU type until it expires. Savings plans cover GPU compute costs only—[storage costs](/pods/storage/types) are billed at standard rates. Storage charges continue to accrue on stopped Pods, and plans are non-refundable with fixed expiration dates.

## Storage pricing

| Storage type | Running Pod | Stopped Pod | Notes |
Expand Down
9 changes: 9 additions & 0 deletions release-notes.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,15 @@ sidebarTitle: "Product updates"
description: "New features, fixes, and improvements for the Runpod platform."
---

<Update label="June 2026">
## Pod deployment experience improvements

- **[Deploy When Available](/pods/choose-a-pod#deploy-when-available):** When your chosen GPU is out of capacity, you can queue your Pod and let Runpod deploy it automatically as soon as the GPU frees up. Set a deployment window using Now, In 24 hours, or a custom range, and you'll get an email and a console inbox notification when your Pod deploys or finishes migrating.
- **[Recommended tab](/pods/choose-a-pod#recommended-tab):** The deploy screen now pairs each template's recommended GPUs with additional high-availability options that are currently in stock.
- **[NVLink availability](/pods/choose-a-pod#nvlink-availability):** When you deploy a multi-GPU Pod on an H100, H200, or B200, a banner shows whether the GPU supports NVLink.

</Update>

<Update label="March 2026">
## Flash beta: Run Python functions on cloud GPUs

Expand Down
Loading