> 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/19.latest/providers/cloudflare.md).

# Cloudflare

Configure Cloudflare Workers AI as a provider for the experimental Extraction capability.

The Cloudflare provider gives access to Clairic, a Cloudflare Workers AI model that reads typed values out of text. The provider supports only the [Extraction](/ai-in-umbraco/19.latest/using-the-api/extraction.md) capability. It has no Chat or Embedding models.

{% hint style="warning" %}
Extraction is **experimental** and disabled by default. Until you enable the `Umbraco:AI:Experimental:Extraction` feature flag, Cloudflare does not appear in the provider list. See [Enabling Extraction](/ai-in-umbraco/19.latest/using-the-api/extraction.md#enabling-extraction).
{% endhint %}

## Installation

{% code title="Package Manager Console" %}

```powershell
Install-Package Umbraco.AI.Cloudflare
```

{% endcode %}

Or via .NET CLI:

{% code title="Terminal" %}

```bash
dotnet add package Umbraco.AI.Cloudflare
```

{% endcode %}

Then enable the feature flag in `appsettings.json` and restart the site:

{% code title="appsettings.json" %}

```json
{
    "Umbraco": {
        "AI": {
            "Experimental": {
                "Extraction": true
            }
        }
    }
}
```

{% endcode %}

## Connection Settings

| Setting    | Required | Description                                                                                                   |
| ---------- | -------- | ------------------------------------------------------------------------------------------------------------- |
| Account ID | Yes      | Your Cloudflare account ID. Find it in the Cloudflare dashboard.                                              |
| API Token  | Yes      | A Cloudflare API token with Workers AI permissions. Stored as a sensitive value and masked in the backoffice. |
| Endpoint   | No       | Custom endpoint URL, for example a proxy. Defaults to `https://api.cloudflare.com/client/v4`.                 |

Both the Account ID and the API Token can reference a configuration value instead of storing it in the database. See [Configuration References](/ai-in-umbraco/19.latest/reference/configuration/ai-options.md#configuration-references). The API Token is a sensitive field, so it can use a `$Umbraco:AI:Secrets:...` reference. The Account ID is not, so use a `$Umbraco:AI:Variables:...` reference for it.

{% code title="Connection settings with configuration references" %}

```
Account ID: $Umbraco:AI:Variables:CloudflareAccountId
API Token:  $Umbraco:AI:Secrets:CloudflareApiToken
```

{% endcode %}

{% hint style="warning" %}
Keep your API token secure. Never commit it to source control or expose it in client-side code.
{% endhint %}

## Models

| Model                    | Name    | Description                                         |
| ------------------------ | ------- | --------------------------------------------------- |
| `@cf/cloudflare/clairic` | Clairic | The only model the provider lists, and the default. |

Create a profile with the **Extraction** capability and the **Clairic** model. Extraction profiles have no extra settings.

Cloudflare bills Workers AI usage to your Cloudflare account. Check Cloudflare's own pricing for Clairic before you use it in production.

## How Requests Are Sent

The provider sends each Extraction request to Cloudflare as one HTTP call, however many fields it holds:

* The request's state is sent as the text to extract from.
* Every field is sent in the same call, keyed by its field ID. The answers come back under the same IDs.
* The reference date, when set, is sent so that relative dates such as "yesterday" resolve correctly.
* When Cloudflare returns a confidence and a probability that the value is not in the text, they are set on the answer.
* Text and list answers include where each value sits in the text. A value that was not found has no span.
* A date and time with a `Z` suffix or a time zone offset, such as `+02:00`, is converted to a UTC `DateTime`.
* Token usage is recorded once for the call.

This means [extracting multiple fields at once](/ai-in-umbraco/19.latest/using-the-api/extraction.md#extracting-multiple-fields-at-once) costs a single request to Cloudflare. The provider sets no limit on the number of fields in a request.

## Test Connection

**Test Connection** searches your Cloudflare account's Workers AI models for Clairic. This checks the Account ID and the API Token without running the model:

* A valid account and token pass the test.
* An invalid token, an unknown account, or any other error fails the test.

A successful check is cached for one hour for the same endpoint, account, and token. Repeated tests within that hour do not send a new request. A failed check is not cached.

## Retries

When Cloudflare responds with `429` (rate limited), the provider retries the request up to 2 more times:

* It waits for the time given in the `Retry-After` header, when present.
* Without the header, it waits 1 second, then 2 seconds.
* A single wait is never longer than 30 seconds.

Other errors are not retried. For example, `401` (invalid token) and `400` (request rejected) fail straight away.

## Related

* [Providers Overview](/ai-in-umbraco/19.latest/providers/providers.md)
* [Extraction](/ai-in-umbraco/19.latest/using-the-api/extraction.md)
* [Managing Connections](/ai-in-umbraco/19.latest/backoffice/managing-connections.md)


---

# 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/19.latest/providers/cloudflare.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.