In App Design Center, you can configure a
[Vertex AI](https://docs.cloud.google.com/vertex-ai/docs/start/introduction-unified-platform)
component to enable Vertex AI APIs and grant connected workloads access
to train and deploy machine learning (ML) models and AI applications. You can
customize the component's connections and configuration parameters based on the
[terraform-google-project-factory](https://github.com/terraform-google-modules/terraform-google-project-factory/tree/main/modules/project_services)
Terraform module.

## Component connections

The following table includes the components that you can connect to a
Vertex AI component, and the resulting updates to your
application and its generated Terraform code.

| Connected component | Application updates | Background information |
|---|---|---|
| [Compute Engine instance template](https://docs.cloud.google.com/application-design-center/docs/configure-compute-engine-instance) | - The Compute Engine instances can interact with Vertex AI services. - The `roles/aiplatform.user` role is added to the Compute Engine instance template service account. | [Vertex AI access control with IAM](https://docs.cloud.google.com/vertex-ai/docs/general/access-control) |
| [Service account](https://docs.cloud.google.com/application-design-center/docs/configure-service-account) | - The service account can interact with Vertex AI services. - The `roles/aiplatform.user` role is added to the service account. | [Vertex AI access control with IAM](https://docs.cloud.google.com/vertex-ai/docs/general/access-control) |
| [Cloud Run](https://docs.cloud.google.com/application-design-center/docs/configure-cloud-run) | - The Cloud Run service can interact with Vertex AI services. - The `roles/aiplatform.user` role is added to the Cloud Run service account. | [Host AI apps and agents on Cloud Run](https://docs.cloud.google.com/run/docs/ai-agents) |

## Required configuration parameters

If your template includes a Vertex AI component, you must configure the
following parameters before you deploy.

| Parameter name | Description and constraints | Background information |
|---|---|---|
| **Project ID** | The project where you want to enable Vertex AI APIs. | [Configure components](https://docs.cloud.google.com/application-design-center/docs/design-application-templates#configure_components) |

## Optional configuration parameters

The following parameters are optional. To display advanced parameters, in the
**Configuration** area, select **Show advanced fields**.

| Feature | Parameter name | Description and constraint information | Background information |
|---|---|---|---|
| <br /> | **Enable APIs** | Enabled by default. To skip API enablement, clear the checkbox. | [Set up a project and a development environment](https://docs.cloud.google.com/vertex-ai/docs/start/cloud-environment) |
| <br /> | **Activate APIs** | By default, the Vertex AI API is activated. If you add other APIs in the *Activate API Identities* section, the additional APIs are appended to this list. | [Service agents](https://docs.cloud.google.com/iam/docs/service-agents) |
| Activate API Identities | **API** | An additional API that you want to activate. A service agent will be created for this API. | [Service agents](https://docs.cloud.google.com/iam/docs/service-agents) |
| Activate API Identities | **Roles** | The roles that you want to grant to the service agent. | [Role list](https://docs.cloud.google.com/iam/docs/roles-permissions) |
| <br /> | **Disable Service on Destroy** | By default, the service is disabled when the Terraform resource is destroyed. | [disable_on_destroy](https://registry.terraform.io/providers/hashicorp/google/latest/docs/resources/google_project_service.html#disable_on_destroy-1) |
| <br /> | **Disable Service Dependent Services** | By default, dependent services are disabled when the Terraform resource is destroyed. | [disable_dependent_services](https://registry.terraform.io/providers/hashicorp/google/latest/docs/resources/google_project_service.html#disable_dependent_services-1) |