-
Notifications
You must be signed in to change notification settings - Fork 18
feat(studio): Add Guardrails and Virtual Model docs #792
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.
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
Merged
htolentino-nvidia
merged 2 commits into
main
from
nvb6478756-add-guardrail-docs/htolentino
Jul 20, 2026
Merged
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,40 @@ | ||
| --- | ||
| title: "NeMo Studio Guardrail Configs" | ||
| description: "" | ||
| --- | ||
| Use **Guardrails** in the NeMo Studio workspace sidebar to view and manage NeMo Guardrails configurations for the selected workspace. | ||
|
|
||
| ## Guardrail Config List | ||
|
|
||
| The Guardrail Configs table shows all guardrail configurations in the current workspace. Select a row to open the config detail page. | ||
|
|
||
| | Action | Where | Result | | ||
| |--------|-------|--------| | ||
| | View details | Select a row | Opens the guardrail config detail page. | | ||
| | Search | Name search bar | Filters the list to configs whose names match the query. | | ||
| | Delete a config | Row actions > **Delete** | Removes the guardrail configuration from the workspace. | | ||
|
|
||
| The table shows each config's name, description, model count, rail count, and last updated time. | ||
|
|
||
| Guardrail configurations are created and updated through the `nemo guardrail` CLI or the Guardrails API. NeMo Studio reflects the current workspace state. | ||
|
|
||
| ## Guardrail Config Detail | ||
|
|
||
| The detail page shows the stored properties of a single guardrail configuration. | ||
|
|
||
| | Field | Description | | ||
| |-------|-------------| | ||
| | Description | A short summary of the config's purpose, if set. | | ||
| | Models | Number of task models (for example, `content_safety`) attached to the config. | | ||
| | Rails | Total count of input, output, and retrieval flows defined in the config. | | ||
| | Created | When the configuration was first stored. | | ||
| | Updated | When the configuration was last modified. | | ||
| | Config | The full configuration payload as formatted JSON. | | ||
|
|
||
| The **Edit** action is not yet available. To modify a configuration, use the `nemo guardrail` CLI or the Guardrails API, then reload the page. | ||
|
|
||
| ## Delete a Guardrail Config | ||
|
|
||
| Select **Delete** from the row actions menu on the list page, or select **Delete** in the detail page header. Confirm the prompt to remove the configuration. Deletion is permanent. | ||
|
|
||
| Before deleting a config, check whether it is referenced by any virtual model middleware pipeline. Removing a config that is wired to a virtual model does not automatically remove the middleware reference. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,47 @@ | ||
| --- | ||
| title: "NeMo Studio Virtual Models" | ||
| description: "" | ||
| --- | ||
| Use **Virtual Models** in the NeMo Studio workspace sidebar to inspect inference routes and their middleware pipelines for the selected workspace. | ||
|
|
||
| ## Virtual Model List | ||
|
|
||
| The Virtual Models table shows all user-created virtual models in the current workspace. Auto-provisioned passthrough routes are hidden. Select a row or choose **View** from the row actions menu to open the details panel. | ||
|
coderabbitai[bot] marked this conversation as resolved.
Outdated
|
||
|
|
||
| | Action | Where | Result | | ||
| |--------|-------|--------| | ||
| | View details | Select a row or row actions > **View** | Opens the virtual model details panel. | | ||
| | Search | Name search bar | Filters the list to virtual models whose names match the query. | | ||
| | Filter by base model | **Default Model** filter | Shows only virtual models backed by the selected base model. | | ||
| | Filter by date | **Created At** filter | Narrows results to a specific creation date range. | | ||
| | Delete a virtual model | Row actions > **Delete** | Removes the virtual model. Requires typing the model name to confirm. | | ||
|
|
||
| The table shows each virtual model's name, default model, middleware call count, and creation time. | ||
|
|
||
| Virtual models are created and updated through the `nemo inference virtual-models` CLI or the Inference Gateway API. NeMo Studio reflects the current workspace state. | ||
|
|
||
| ## Virtual Model Details | ||
|
|
||
| Selecting a row opens a details panel with the full configuration of a single virtual model. | ||
|
|
||
| ### Summary | ||
|
|
||
| | Field | Description | | ||
| |-------|-------------| | ||
| | Created | When the virtual model was first created. | | ||
| | Default model | The backend model entity that the Inference Gateway proxies requests to. | | ||
| | Autoprovisioned | Whether the route was created automatically by the platform. Auto-provisioned models are hidden from the list. | | ||
| | Override proxy | An optional plugin-provided proxy identifier (format: `plugin-name.proxy-name`) that overrides the default IGW proxy, if set. | | ||
| | Models | Backend model entries associated with this virtual model, displayed as `model (backend_format)`. | | ||
|
|
||
| ### Middleware | ||
|
|
||
| The details panel shows three ordered middleware pipelines applied to every request through this virtual model. | ||
|
|
||
| | Pipeline | When it runs | | ||
| |----------|-------------| | ||
| | Request middleware | Before the request reaches the backend model. | | ||
| | Response middleware | After the model responds, before the response is returned to the caller. | | ||
| | Post-response middleware | After the response is returned to the caller. | | ||
|
|
||
| Each middleware entry shows its plugin name, config type, config reference, and inline config payload if one is set. A guardrails middleware entry, for example, shows the plugin `nemo-guardrails`, the config type `guardrail_config`, and a config ID pointing to a workspace guardrail configuration. | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.