Google Vertex AI
Before you begin
Before starting, you'll need:
- The JSON key token for your GCP service account.
- Your GCP service account's email address.
If you don't have a GCP service account created yet, check out our guide How to Create a GCP Service Account for use with Google Vertex AI.
Configure the AI integration
- Click AI Platforms from the main menu.
- Click New AI Platform.
- Select Google Vertex AI.
- Enter the name, your service account's JSON key, and select your region using the Vertex AI Parameters table below as a guide, then click Create Google Vertex AI Integration.
Vertex AI Parameters
Field | Description | Required |
---|---|---|
Name | A descriptive name to identify the integration within Vectorize. | Yes |
Service Account JSON | The JSON key token for your GCP service account. | Yes |
Region | The Google region to use. | Yes |
Configuring the Vertex AI integration in a RAG Pipeline
You can think of the Vertex AI integration as having two parts to it. The first is authorization for your region. This part is re-usable across pipelines and allows you to connect to this same integration in different pipelines without providing the credentials and specifying the region every time.
The second part is the configuration that's specific to your RAG Pipeline, which allows you to specify which foundation model to use.
What's next?
-
If you haven't yet built a connector to your source database, go to configure a source connector and select your source data platform.
OR
-
If you're ready to start creating pipelines for vectorizing data, head to Pipeline Basics.