docs: Document Pod deploy-flow updates and discontinue savings plans in docs#672
docs: Document Pod deploy-flow updates and discontinue savings plans in docs#672promptless[bot] wants to merge 1 commit into
Conversation
- choose-a-pod: add Recommended GPU tab, NVLink banner, and Deploy when available sections (CON-34, CON-9, CON-132/152/10) - pricing/overview: mark savings plans as discontinued; existing plans run to expiration (CON-171) - manage-pods: add Report a problem with a template note (CON-108) Ref: DOCS-450
![promptless[bot]](https://avatars.githubusercontent.com/in/979718?s=60&v=4){kind=small}
|
|
||
| <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 and can no longer be created. Existing savings plans are unaffected and continue to run until their term expires. |
There was a problem hiding this comment.
Reframed the Savings plans section as discontinued because CON-171 hides savings plan creation entry points and notes ~300 affected users keep their active plans until they expire (sunset after 60 days). Replaced the prior "commit to a term" creation copy in pricing.mdx.
Source: https://linear.app/runpod/issue/CON-171/discontinue-savings-plans
|
|
||
| When you deploy a Pod from the [console](https://www.console.runpod.io/pods), the GPU selector groups available hardware into tabs to help you find a suitable option: | ||
|
|
||
| - **Recommended**: Shows GPUs recommended for your selected template, followed by additional high-availability GPUs that are in stock. Use this tab when you want a configuration that's known to work with your template and is available to deploy right now. |
There was a problem hiding this comment.
Documented the Recommended tab behavior from CON-34: it shows the template's recommended GPUs plus the top available high-availability GPUs (stockStatus not null). Aligned the tab descriptions with the existing get-started.mdx deploy flow.
Source: https://linear.app/runpod/issue/CON-34/show-high-availability-options-in-recommended
|
|
||
| ### NVLink for multi-GPU Pods | ||
|
|
||
| When you deploy a Pod with more than one GPU, the console displays an NVLink banner for GPU types that support it. NVLink provides high-bandwidth, direct GPU-to-GPU communication, which improves performance for multi-GPU workloads such as distributed training and large-model inference. Runpod's H100, H200, and B200 SKUs support NVLink. If you select a multi-GPU configuration on a GPU type without NVLink, no banner appears. |
There was a problem hiding this comment.
Documented the NVLink banner for multi-GPU deploys from CON-9. The acceptance criteria state the hardcoded banner shows for H100, H200, and B200 (SXM/NVL cards), which is why those three SKUs are listed.
Source: https://linear.app/runpod/issue/CON-9/show-nvlink-availability-for-multi-gpu-pod-deploys
|
|
||
| ### Deploy when a GPU is available | ||
|
|
||
| If the GPU you want is out of capacity, you don't have to wait and retry manually. Select the out-of-capacity GPU and choose **Deploy when available** instead of deploying immediately. Runpod monitors for capacity and deploys your Pod automatically when that GPU frees up within the time window you set, then notifies you by email and in the console inbox. |
There was a problem hiding this comment.
Summarized Deploy when available (CON-132/CON-152/CON-10): out-of-capacity GPUs can auto-deploy within a window that defaults to 24 hours with hour/minute precision, and email + console inbox notifications fire on deploy. Cross-linked to the full modal flow already in get-started.mdx rather than duplicating it.
Source: https://linear.app/runpod/issue/CON-152/update-deploy-when-available-from-and-to-ux
|
|
||
| ## Report a problem with a template | ||
|
|
||
| When you open a Pod's details, the **Template details** section shows information about the template the Pod is running. 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 to the Runpod team to help improve the template. |
There was a problem hiding this comment.
Added the Template details / Report problem with template note from CON-108, which renames the Pod inspector's "Template readme" section to "Template details" and adds a button opening an open-ended feedback modal (routed to a PostHog survey).
Source: https://linear.app/runpod/issue/CON-108/add-template-feedback-inside-the-pod-details
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
Open this suggestion in Promptless to view citations and reasoning process
Updates Pod docs for Console Experience changes (DOCS-450): marks savings plans as discontinued in pricing.mdx and overview.mdx, documents the Recommended GPU tab, the NVLink banner for multi-GPU deploys, and Deploy When Available on choose-a-pod.mdx, and adds a "Report a problem with a template" note to manage-pods.mdx.
Trigger Events
Tip: Tag @Promptless in a GitHub issue to turn it into a documentation update 🐙