> For the complete documentation index, see [llms.txt](https://docs.umbraco.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.umbraco.com/ai-in-umbraco/backoffice/managing-profiles.md).

# Managing Profiles

Create and manage AI profiles in the Umbraco backoffice.

Profiles combine a connection with model settings for specific use cases. Create different profiles for different AI tasks.

## Viewing Profiles

1. Navigate to the **AI** section > **Profiles**.
2. The profile list shows all configured profiles.
3. Use the filter field to search profiles by name. The **Capability** column shows each profile's capability, such as Chat, Embedding, Speech-to-Text, or Image Generation and Decision when enabled.
4. Click a profile to view or edit its details.

![The profiles list showing the Default Chat profile](https://583970477-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCHKT9bQhbR3swAdBvil4%2Fuploads%2Fgit-blob-4b7895074be239074113c05eea9efc454ff9c1f0%2Fbackoffice-profiles-list.png?alt=media)

## Creating a Profile

1. Click **Create Profile** in the Profiles section.
2. Fill in the required fields:

| Field          | Description                                                                                               |
| -------------- | --------------------------------------------------------------------------------------------------------- |
| **Name**       | Display name for the profile                                                                              |
| **Alias**      | Unique identifier (used in code and as default)                                                           |
| **Capability** | Type of AI operation (Chat, Embedding, Speech-to-Text, or the experimental Image Generation and Decision) |
| **Connection** | Which connection to use                                                                                   |
| **Model**      | The specific AI model                                                                                     |

3. Configure capability-specific settings.
4. Click **Save**.

![The Create Profile capability selection](https://583970477-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCHKT9bQhbR3swAdBvil4%2Fuploads%2Fgit-blob-cdca51504be7d59c2c89f6e388f1155321c32a6f%2Fbackoffice-create-profile-modal.png?alt=media)

![The Create Profile form with connection and model settings](https://583970477-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCHKT9bQhbR3swAdBvil4%2Fuploads%2Fgit-blob-d76e493e470969f0b54f0405f7eab576568aa09b%2Fbackoffice-create-profile-form.png?alt=media)

## Chat Profile Settings

![Chat profile settings showing Temperature, Max Tokens, and System Prompt](https://583970477-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCHKT9bQhbR3swAdBvil4%2Fuploads%2Fgit-blob-78437e37c3c7e268fbd876869b2aeb55fc3b5771%2Fbackoffice-chat-profile-settings.png?alt=media)

For profiles with the **Chat** capability:

| Setting           | Description                                                             | Default       |
| ----------------- | ----------------------------------------------------------------------- | ------------- |
| **Temperature**   | Controls randomness (0-2). Lower = more focused, higher = more creative | Model default |
| **Max Tokens**    | Maximum tokens in the response                                          | Model default |
| **System Prompt** | Instructions sent with every request                                    | None          |

### Temperature Guidelines

| Value     | Best For                                            |
| --------- | --------------------------------------------------- |
| 0.0 - 0.3 | Factual responses, code generation, data extraction |
| 0.4 - 0.7 | Balanced responses, general assistance              |
| 0.8 - 1.2 | Creative writing, brainstorming                     |

### System Prompts

System prompts set the AI's behavior for all requests using this profile:

{% code title="System Prompt Example" %}

```
You are a helpful content editor for a corporate website.
Write in a professional tone. Keep responses concise.
Always suggest improvements rather than making changes without explanation.
```

{% endcode %}

{% hint style="info" %}
Leave settings empty to use the model's default values. Only configure what you need to override.
{% endhint %}

## Embedding Profile Settings

Embedding profiles currently use model defaults. Select the appropriate embedding model for your use case:

| Model                  | Dimensions | Best For                        |
| ---------------------- | ---------- | ------------------------------- |
| text-embedding-3-small | 1536       | Cost-effective, general purpose |
| text-embedding-3-large | 3072       | Higher accuracy, more storage   |

## Speech-to-Text Profile Settings

For profiles with the **Speech-to-Text** capability:

| Setting      | Description                                                                      | Default       |
| ------------ | -------------------------------------------------------------------------------- | ------------- |
| **Language** | Optional BCP-47 language hint for transcription (for example, `en`, `de`, `ja`). | Model default |

{% hint style="info" %}
Leave the language empty to let the model auto-detect the spoken language.
{% endhint %}

## Image Generation Profile Settings

{% hint style="warning" %}
Image Generation is experimental. The **Image Generation** capability can only be selected when the `Umbraco:AI:Experimental:ImageGeneration` feature flag is enabled. See [Enabling image generation](/ai-in-umbraco/using-the-api/image-generation.md#enabling-image-generation).
{% endhint %}

For profiles with the **Image Generation** capability:

| Setting        | Description                                                                                                   | Default          |
| -------------- | ------------------------------------------------------------------------------------------------------------- | ---------------- |
| **Size**       | Default image size as `{width}x{height}`, for example `1024x1024`. A dropdown when the model lists its sizes. | Provider default |
| **Media Type** | Output image format, for example `image/png`, `image/jpeg`, or `image/webp`. Supported values vary by model.  | Provider default |

{% hint style="info" %}
The **Media Type** field is hidden for models that don't support it. Image count and response format are set per request, not on the profile.
{% endhint %}

## Decision Profile Settings

{% hint style="warning" %}
Decision is experimental. The **Decision** capability can only be selected when the `Umbraco:AI:Experimental:Decision` feature flag is enabled. See [Enabling Decision](/ai-in-umbraco/using-the-api/decision.md#enabling-decision).
{% endhint %}

Decision profiles have no extra settings. The settings view shows a message saying so. Select a connection that supports Decision, such as [TypeSafe AI](/ai-in-umbraco/providers/typesafe.md), and its model.

## Governance

Chat profiles have a **Governance** tab where you can assign [guardrails](/ai-in-umbraco/backoffice/managing-guardrails.md) to enforce safety, compliance, and quality rules on all AI operations using that profile.

1. Open a chat profile.
2. Go to the **Governance** tab.
3. Click **Add Guardrail** to select guardrails.
4. Save the profile.

![The Governance tab showing assigned guardrails](https://583970477-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FCHKT9bQhbR3swAdBvil4%2Fuploads%2Fgit-blob-34329c87167a387a46a0bc8864b374722c33f688%2Fbackoffice-profile-governance-tab.png?alt=media)

{% hint style="info" %}
The Governance tab is only available for **Chat** profiles. Embedding profiles do not support guardrails.
{% endhint %}

## Setting Default Profiles

To use a profile as the default when no profile is specified in code:

1. Navigate to the **AI** section > **Settings**.
2. Select your chat profile via the **Default Chat Profile** picker.
3. Select your embedding profile via the **Default Embedding Profile** picker.
4. Select your speech-to-text profile via the **Default Speech-to-Text Profile** picker.
5. If Image Generation is enabled, select your Image Generation profile via the **Default Image Generation Profile** picker.
6. If Decision is enabled, select your Decision profile via the **Default Decision Profile** picker.
7. Click **Save**.

{% hint style="info" %}
See [Managing Settings](/ai-in-umbraco/backoffice/managing-settings.md) for more details on configuring default profiles.
{% endhint %}

## Editing a Profile

1. Click on the profile in the list.
2. Modify the desired fields.
3. Click **Save**.

{% hint style="warning" %}
Changing a profile's capability is not supported after creation. Create a new profile instead.
{% endhint %}

## Deleting a Profile

1. Open the profile you want to delete.
2. Click **Delete** in the actions menu.
3. Confirm the deletion.

{% hint style="warning" %}
Ensure no code depends on this profile before deleting. Check for usages of the profile alias in your codebase.
{% endhint %}

## Using Tags

Tags help organize profiles:

1. Add tags when creating or editing a profile.
2. Use tags to categorize by team, project, or purpose.
3. Tags are visible in the profile list.

Example tags: `production`, `content-team`, `experimental`

## Related

* [Managing Connections](/ai-in-umbraco/backoffice/managing-connections.md) - Connections are required for profiles
* [Profiles Concept](/ai-in-umbraco/concepts/profiles.md) - Deeper explanation of profiles
* [Using Profiles in Code](/ai-in-umbraco/using-the-api/chat/basic-chat.md) - How to use profiles programmatically


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.umbraco.com/ai-in-umbraco/backoffice/managing-profiles.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.