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

# Provider configuration

> Centrally configure AI inference providers for your organization so developers don't need individual API keys.

Cline Enterprise lets administrators configure inference providers once from the Cline admin console. All organization members automatically receive the provider settings when they sign in — no individual API key distribution required.

## How it works

<CardGroup cols={2}>
  <Card title="Admin configures once" icon="gear">
    The admin selects a provider, enters credentials and settings, and saves the configuration in the Cline admin console.
  </Card>

  <Card title="Automatic distribution" icon="shield-check">
    Configuration is pushed to all signed-in organization members automatically. Members cannot switch to other providers or their personal Cline accounts.
  </Card>

  <Card title="Simplified member onboarding" icon="user-plus">
    New members get access to the configured provider immediately after signing in with SSO — no setup required on their part.
  </Card>

  <Card title="Consistent experience" icon="users">
    Every developer in the organization uses the same models, regions, and settings.
  </Card>
</CardGroup>

<Warning>
  Once remote provider configuration is enabled and saved, organization members cannot switch to personal Cline accounts or select other providers. Ensure the configured provider is accessible to all members before saving.
</Warning>

## Accessing the admin console

1. Sign in at [app.cline.bot](https://app.cline.bot) with an Admin or Owner account.
2. Go to **Settings → Cline Settings**.
3. Toggle on **Enable settings** to reveal the provider configuration options.

## Supported providers

| Provider             | Best for                                     | Member credential setup                                |
| :------------------- | :------------------------------------------- | :----------------------------------------------------- |
| **Amazon Bedrock**   | Organizations using AWS infrastructure       | AWS credentials in VS Code                             |
| **LiteLLM proxy**    | Multi-model access via a unified proxy       | API key or endpoint (or centralized with a master key) |
| **Google Vertex AI** | Organizations using Google Cloud Platform    | Service account or ADC credentials in VS Code          |
| **Cline (native)**   | Organizations using Cline's managed provider | No individual keys — fully managed                     |

## Provider setup

<Tabs>
  <Tab title="Amazon Bedrock">
    ### Before you begin

    * Administrator access to the Cline admin console (verify by navigating to **Settings → Cline Settings**)
    * An AWS account with Amazon Bedrock enabled and appropriate permissions
    * Your preferred AWS region (check model availability before choosing)

    ### Admin configuration steps

    <Steps>
      <Step title="Select the provider">
        In **Settings → Cline Settings**, open the **API Provider** dropdown and select **Amazon Bedrock**.
      </Step>

      <Step title="Configure Bedrock settings">
        <AccordionGroup>
          <Accordion title="Region (required)">
            Enter your AWS region, for example `us-east-1` or `us-west-2`. This region is enforced for all organization members.

            <Tip>
              `us-east-1` and `us-west-2` have the broadest model availability. Check the [AWS Global Infrastructure page](https://aws.amazon.com/about-aws/global-infrastructure/regions_az/) if you need a specific region.
            </Tip>
          </Accordion>

          <Accordion title="Custom VPC endpoint (optional)">
            If your organization uses a private VPC endpoint for Bedrock, specify it here to route all API calls through your network perimeter.

            See [AWS PrivateLink for Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/usingVPC.html) for setup instructions.
          </Accordion>

          <Accordion title="Cross-region inference (optional)">
            Enable this to allow Bedrock to automatically route requests to other regions when your primary region has capacity constraints. Useful for maintaining availability during high-demand periods.

            See [Inference Profiles](https://docs.aws.amazon.com/bedrock/latest/userguide/inference-profiles-support.html) for details.
          </Accordion>

          <Accordion title="Global inference profile (optional)">
            Routes requests to the optimal AWS region globally based on availability and latency.
          </Accordion>

          <Accordion title="Prompt caching (optional)">
            Caches consistent portions of prompts across requests, reducing latency and cost for repeated interactions.

            See [Prompt caching in Bedrock](https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html).
          </Accordion>
        </AccordionGroup>
      </Step>

      <Step title="Save the configuration">
        Close the provider configuration panel and click **Save** on the settings page. All signed-in organization members begin using Amazon Bedrock with your settings immediately.

        <Note>
          Click **Save** on the main settings page — closing the configuration panel alone does not save.
        </Note>
      </Step>
    </Steps>

    ### What members need to do

    After the admin saves the configuration, each developer must add their AWS credentials to the Cline extension in VS Code. They can use:

    * An AWS profile configured with `aws configure`
    * Environment variables (`AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, `AWS_REGION`)
    * An IAM role (if running in an AWS-managed environment)

    Members do not choose the region or provider — those are enforced by the admin configuration.

    ### Verification

    1. Confirm the provider shows as **Amazon Bedrock** in the Enabled provider field.
    2. Verify the settings persist after refreshing the page.
    3. Sign in with a member account and confirm only Bedrock is available as a provider.

    For further details, see the [AWS Bedrock documentation](https://docs.aws.amazon.com/bedrock/latest/userguide/what-is-bedrock.html).
  </Tab>

  <Tab title="LiteLLM">
    ### Before you begin

    * Administrator access to the Cline admin console
    * A running LiteLLM proxy instance accessible from your team's development environments
    * The base URL of your LiteLLM proxy
    * Optionally, a master key if your proxy requires authentication

    <Note>
      If you have not deployed a LiteLLM proxy yet, work with your infrastructure team to stand one up first. See the [LiteLLM documentation](https://docs.litellm.ai/) for deployment options.
    </Note>

    ### Admin configuration steps

    <Steps>
      <Step title="Select the provider">
        In **Settings → Cline Settings**, open the **API Provider** dropdown and select **LiteLLM**.
      </Step>

      <Step title="Configure LiteLLM settings">
        <AccordionGroup>
          <Accordion title="Base URL (required)">
            Enter the full URL of your LiteLLM proxy, for example `https://litellm.yourcompany.com` or `http://your-proxy:4000`.

            <Tip>
              Use HTTPS in production. Verify the URL is reachable from all developer machines and CI environments that will use Cline.
            </Tip>
          </Accordion>

          <Accordion title="Master key (optional)">
            If your LiteLLM proxy requires authentication, enter the master key here. This key is used to authenticate all requests from organization members — members do not need to manage individual keys.

            <Warning>
              The master key grants full access to your LiteLLM proxy. Store it securely and rotate it according to your organization's key management policies.
            </Warning>
          </Accordion>
        </AccordionGroup>
      </Step>

      <Step title="Save the configuration">
        Close the provider configuration panel and click **Save** on the settings page. Organization members receive the LiteLLM configuration automatically.

        <Note>
          Click **Save** on the main settings page — closing the configuration panel alone does not save.
        </Note>
      </Step>
    </Steps>

    ### What members need to do

    If a master key is configured at the organization level, members need no additional credentials — access is fully managed. If no master key is configured, members add their individual LiteLLM API key in the Cline VS Code extension.

    ### Verification

    1. Confirm the provider shows as **LiteLLM** in the Enabled provider field.
    2. Verify settings persist after refreshing the page.
    3. Sign in with a member account and confirm the configured models are available.

    ### Troubleshooting

    | Symptom                        | Fix                                                                                      |
    | :----------------------------- | :--------------------------------------------------------------------------------------- |
    | Members don't see the provider | Verify you clicked Save. Confirm the member's account belongs to your organization.      |
    | Connection errors              | Verify the base URL is correct and reachable. Check firewall rules for member IP ranges. |
    | Authentication failures        | Verify the master key is correctly entered and has the right permissions in LiteLLM.     |
    | Models not available           | Confirm the models are configured in your LiteLLM proxy deployment.                      |

    For further details, see the [LiteLLM documentation](https://docs.litellm.ai/).
  </Tab>

  <Tab title="Google Vertex AI">
    ### Before you begin

    * Administrator access to the Cline admin console
    * A Google Cloud project with Vertex AI enabled
    * Your GCP project ID and preferred region

    ### Admin configuration steps

    <Steps>
      <Step title="Select the provider">
        In **Settings → Cline Settings**, open the **API Provider** dropdown and select **Google Vertex AI**.
      </Step>

      <Step title="Configure Vertex AI settings">
        Enter your **GCP Project ID** and select your **region**. These settings are enforced for all organization members.
      </Step>

      <Step title="Save the configuration">
        Close the provider configuration panel and click **Save**. Organization members receive the Vertex AI configuration automatically.
      </Step>
    </Steps>

    ### What members need to do

    Each developer must configure Google Cloud credentials on their machine. They can use:

    * Application Default Credentials (ADC) via `gcloud auth application-default login`
    * A service account key file
    * Workload identity (if running on GCP)

    For further details, see the [Vertex AI documentation](https://cloud.google.com/vertex-ai/docs).
  </Tab>
</Tabs>

## Updating provider settings

You can change provider settings at any time from **Settings → Cline Settings**. Changes take effect for all organization members immediately after you save.

To switch to a different provider entirely, select the new provider from the **API Provider** dropdown, complete the configuration, and save. Notify your team that they may need to configure new credentials locally.