Please use the form below to provide your feedback. Because your feedback is valuable to us, the information you submit in this form is recorded in our issue tracking system (JIRA), which is publicly available. You can track the status of your feedback using the ticket number displayed in the dialog once you submit the form.

Configure Capella iQ

  • Capella Operational
  • how-to
Capella iQ harnesses the power of a third-party large language model (LLM) to improve your queries. You can turn this tool on or off for your organization.

Prerequisites

Accept Capella iQ Supplemental Terms

Before anyone in an organization can use Capella iQ, an Organization Owner must accept the Capella Supplemental Terms. If the terms are not yet accepted, the iQ button is still visible in the query editor. If the supplemental terms have not been accepted and a user without the Organization Owner role clicks the iQ button, they’re informed that Capella iQ is disabled. If a user with the Organization Owner role clicks the iQ button, they can instead review and accept the terms for the current organization.

To accept the Capella iQ supplemental terms:

  1. In the navigation breadcrumbs in the Capella UI, do 1 of the following:

    • Click your organization name and go to Operational.

    • Click your current project name or search for a project and go to Operational.

    • Expand the cluster breadcrumb and search for a cluster.

  2. Select the cluster where you want to configure Capella iQ.

  3. Click Data Tools  Query.

  4. Click iQ.

  5. In the prompt, review and accept the Capella iQ Supplemental Terms.

  6. Click Continue.

    Capella iQ is now available to all your organization’s users, who can use it with all of your organization’s clusters.

Accepting the terms for Capella iQ enables iQ for both operational clusters and Capella Analytics clusters. While iQ helps you write SQL++ queries in an operational cluster, iQ for Capella Analytics helps you write SQL++ for Capella Analytics queries and statements.

Select Your Capella iQ Model Provider

Capella iQ supports 2 model providers:

Model provider selections made at the organization level are automatically applied at the cluster level. For example, if the Organization Owner enables only OpenAI, all current and new clusters would use OpenAI as their model provider.

If your organization enables both model providers, the default model provider assigned to a cluster depends on the cluster’s deployment location. Users can also choose which available model provider to use for a specific cluster.

The model provider you select in iQ also applies to iQ Insights, ensuring consistent responses across both services.

Select Your Organization’s Capella iQ Model Provider

Organization Owners can make 1 or both model providers (OpenAI and AWS Bedrock) available to your organization. If both providers are enabled, anyone in the organization with the Cluster Manager or Cluster Data Reader/Writer role can choose their preferred model provider at the cluster level directly in the Capella iQ panel.

To select the model providers for Capella iQ in your organization:

  1. In the navigation breadcrumbs in the Capella UI, click your organization name.

  2. Go to Settings.

  3. In Capella iQ, under Model Provider, select 1 or both of the available model providers:

    • OpenAI

    • AWS Bedrock

  4. Click Save.

    The selected model providers are now available to all clusters in the organization.

Set Your Preferred Model Provider

If the Organization Owner has enabled both model providers (OpenAI and AWS Bedrock), users with the Cluster Manager or Cluster Data Reader/Writer role can choose their preferred model provider directly in the Capella iQ panel for a specific cluster. If only 1 model provider is enabled at the organization level, all clusters must use that model for all iQ actions.

Capella saves the model provider selection per user. Each user on the same cluster can have a different model provider selected. For example, 1 user can use AWS Bedrock while another uses OpenAI on the same cluster.

To select the model provider for Capella iQ at the cluster level:

  1. In the navigation breadcrumbs in the Capella UI, do 1 of the following:

    • Click your organization name and go to Operational.

    • Click your current project name or search for a project and go to Operational.

    • Expand the cluster breadcrumb and search for a cluster.

  2. Select the cluster where you want to configure Capella iQ.

  3. Click Data Tools  Query.

  4. Click iQ.

  5. In the iQ Chat window, select your preferred model provider for this cluster:

    • OpenAI

    • AWS Bedrock

    Capella iQ uses your selected provider for all of your subsequent queries on this cluster.

Turn Capella iQ Off for Your Organization

Capella iQ is on by default and you control it at the organization level. To turn Capella iQ on or off for your organization, you need to navigate to your organization’s general settings page:

  1. In the navigation breadcrumbs in the Capella UI, click your organization name.

  2. Go to Settings  General.

  3. Select or deselect Enable Capella iQ and click Save.

    If Capella iQ is turned off for an organization, the iQ button is still visible in the query editor. Any user who clicks this button sees a message that Capella iQ is disabled.

When you disable Capella iQ, you also disable iQ Insights.

Next Steps