Skip to main content

Amazon Bedrock Integration

Amazon Bedrock Setup​

Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models from leading AI companies. To use Amazon Bedrock, you need an AWS account with Bedrock access. Follow these steps to set up your AWS credentials:

  1. Sign in to the AWS Management Console and open the IAM console at https://console.aws.amazon.com/iam/.
  2. In the navigation pane on the left, choose Users. This will take you to the list of IAM users in your account.
  3. Click the Create user button to start the process of creating a new IAM user.
  4. For User name, enter a name for the user (e.g., "bedrock-api-user"). Choose a name that clearly indicates the purpose of this user.
  5. Leave the "Provide user access to the AWS Management Console" checkbox unchecked. We only need programmatic access for Bedrock API interactions.
  6. Click Next to proceed to the permissions page.
  7. On the Set permissions page, select Attach existing policies directly.
  8. In the search box, type Bedrock to filter the policies. Select the checkbox next to AmazonBedrockFullAccess. This policy grants the necessary permissions to interact with Amazon Bedrock services.
  9. Click Next to review the user details and permissions.
  10. Review the details to ensure everything is correct, then click Create user to finalize the user creation.
  11. You'll see a success message indicating that the user has been created. Now we need to create access keys for this user to enable programmatic access.
  12. From the list of users, click on the name of the user you just created to access their details page.
  13. Go to the Security credentials tab. This is where we manage the user's access keys and other security settings.
  14. In the Access keys section, click Create access key. Access keys are necessary for programmatic access to AWS services.
  15. On the Select access key use case page, choose Third-party service. This is the most appropriate option for accessing Bedrock through an external application or service.
  16. Click Next to proceed to the optional tagging step.
  17. (Optional) You can add a description tag for the access key if desired. This can be helpful for tracking the purpose or usage of this specific key.
  18. Click Create access key to generate the new access key pair.
  19. On the final page, you'll see the Access key ID and Secret access key. This is the only time you'll be able to view the Secret access key, so it's crucial to save it securely now.
  20. Download the .csv file containing the credentials or copy both the Access key ID and Secret access key to a secure location.
  21. Click Done to complete the process.

Important: Keep these credentials secure. Do not share them publicly or commit them to version control systems.

When setting up your integration, we recommend using either the us-east-1 or us-west-2 regions, as these contain the most models. For specific model availability by region, please refer to the AWS documentation.

Enabling Models in Amazon Bedrock​

To use models through Amazon Bedrock, you need to enable them. Here's how:

  1. Ensure you have the necessary permissions to request or modify access to Amazon Bedrock foundation models.

  2. Sign in to the Amazon Bedrock console at https://console.aws.amazon.com/bedrock/.

  3. In the left navigation pane, under Bedrock configurations, choose Model access.

  4. On the Model access page, click Modify model access.

  5. Select the models you want to enable for your account:

    • Check the box next to individual models to select or deselect them.
    • Use the top checkbox to select or deselect all models at once.
    • You can group models (e.g., by provider) and select/deselect all models in a group.

    Important: Review the End User License Agreement (EULA) for each model before requesting access.

  6. After making your selections, click Next.

  7. If you've selected any Anthropic models, you'll need to provide use case details:

    • Click Submit use case details.
    • Fill out the form describing your intended use of the models.
    • Click Submit form.

    Note: Access to Anthropic models is granted or denied based on your responses.

  8. Review your access change requests and carefully read the Terms.

    Note: Your use of Amazon Bedrock foundation models is subject to the seller's pricing terms, EULA, and the AWS service terms.

  9. If you agree to the terms, click Submit.

  10. The changes may take several minutes to be reflected in the console.

    Note: If you're revoking access to a model, it may still be accessible via API for a short time while changes propagate. To immediately revoke access, you can add an IAM policy to deny access to the specific model.

  11. If your request is successful, the Access status for each model will change to either "Access granted" or "Available to request".

  12. If you encounter an error indicating lack of permissions, contact your account administrator to request access or to grant you the necessary permissions.

Important: Enabling a model may have associated costs. Always review the pricing details before enabling a model.