> ## Documentation Index
> Fetch the complete documentation index at: https://docs.requesty.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Managed Policies

> Routing policies maintained by Requesty for the best experience out of the box

Managed Policies are routing policies that Requesty creates and maintains for you. They combine the best providers for a model into a single ready-to-use policy, so you get reliable routing without building or maintaining a policy yourself.

<Note>
  **[Browse managed policies](https://app.requesty.ai/routing-policies?tab=managed)** in the Requesty Console.
</Note>

## How It Works

<Steps>
  <Step title="Requesty maintains the policy">
    Requesty curates the model chain behind each managed policy and keeps it up to date as providers change. You never need to edit anything.
  </Step>

  <Step title="You reference the policy by name">
    Use the policy name as-is as the `model` in your request. Unlike custom policies, managed policies do not use the `policy/` prefix.
  </Step>

  <Step title="The router does the rest">
    Requests are routed across the models in the policy. Models that are not on your approved list are skipped automatically.
  </Step>
</Steps>

## Benefits

<CardGroup cols={2}>
  <Card title="Zero Setup" icon="wand-magic-sparkles">
    Ready-made policies with no configuration required.
  </Card>

  <Card title="Maintained by Requesty" icon="shield-check">
    We update the model chain as providers evolve, so you always route on current best practice.
  </Card>

  <Card title="High Availability" icon="check-circle">
    Multiple providers behind one name means fewer failed requests.
  </Card>

  <Card title="Override When Needed" icon="sliders">
    Create a custom policy with the same name and yours takes precedence.
  </Card>
</CardGroup>

## Using a Managed Policy

<Steps>
  <Step title="Find a Managed Policy">
    Go to the [Managed tab](https://app.requesty.ai/routing-policies?tab=managed) on the Routing Policies page. Each row shows the policy name, its copyable model ID, and the models it routes across. Click a policy to see its full model chain.
  </Step>

  <Step title="Use the Policy in Your Code">
    Reference the policy name directly as the `model`. No `policy/` prefix:

    <CodeGroup>
      ```python Python theme={"dark"}
      from openai import OpenAI

      client = OpenAI(
          base_url="https://router.requesty.ai/v1",
          api_key="your-requesty-api-key"
      )

      response = client.chat.completions.create(
          model="managed-policy-name",
          messages=[{"role": "user", "content": "Hello!"}]
      )
      ```

      ```typescript TypeScript theme={"dark"}
      import OpenAI from 'openai';

      const client = new OpenAI({
        baseURL: 'https://router.requesty.ai/v1',
        apiKey: 'your-requesty-api-key'
      });

      const response = await client.chat.completions.create({
        model: 'managed-policy-name',
        messages: [{ role: 'user', content: 'Hello!' }]
      });
      ```

      ```bash cURL theme={"dark"}
      curl https://router.requesty.ai/v1/chat/completions \
        -H "Content-Type: application/json" \
        -H "Authorization: Bearer your-requesty-api-key" \
        -d '{
          "model": "managed-policy-name",
          "messages": [{"role": "user", "content": "Hello!"}]
        }'
      ```
    </CodeGroup>
  </Step>
</Steps>

## Managed vs Custom Policies

|                     | Managed Policies    | Custom Policies           |
| ------------------- | ------------------- | ------------------------- |
| Created by          | Requesty            | You                       |
| Maintenance         | Automatic           | Manual                    |
| Model reference     | Policy name as-is   | `policy/your-policy-name` |
| Strategy and models | Curated by Requesty | Fully configurable        |

<Note>
  If your organization has a custom policy with the same name as a managed policy, your custom policy takes precedence.
</Note>

## Good to Know

* Models in a managed policy that are not on your organization's approved model list are skipped automatically. They appear grayed out with a warning in the Console.
* Managed policies work with the same analytics, cost tracking, and logs as any other request.
* Want full control over strategy and models? Create your own [Fallback](/features/fallback-policies), [Load Balancing](/features/load-balancing-policies), or [Latency](/features/latency-routing) policy instead.