Skip to main content

DBT Quick Start

Goal of this guide

Create a new Wren AI project from your local dbt project and learn how to keep it in sync.

plan support

DBT Integration is only available in Essential/ Business Plan and above

supported databases

Supported Datasource

  • PostgreSQL
  • MySQL
  • BigQuery
  • Redshift
  • Snowflake

Use the Wren CLI to import your dbt project into Wren AI Cloud. We sync your analytics-ready dbt models (e.g., marts) — including model, column descriptions and relationships — as Models, ready for natural-language querying.

In this quick start, you'll:

  • Install the CLI
  • Log in to Wren AI Cloud
  • Prepare your dbt project (dbt build + dbt docs generate)
  • Sync your dbt models to Wren AI
  • Optionally push updates later with wren dbt update

1. Install the CLI​

Follow the same installation steps as in the Installation and Login guide

If you already have the CLI installed, verify your version:

wren version

2. Log in to Wren AI Cloud​

Follow the same login steps as in the Installation and Login guide If you already have the CLI installed and logged in, verify your status:

wren auth status

3. Prepare your dbt project​

From your dbt project directory:

cd /path/to/your-dbt-project

(1) Build your dbt project​

dbt build

This generates the manifest.json artifact, which contains a complete representation of your dbt project's resources (including model descriptions) that Wren AI uses to understand your models. See dbt's manifest docs: Manifest JSON file.

(2) Generate dbt documentation​

dbt docs generate

This creates the required catalog.json file in your target/ directory. The catalog.json includes column types that Wren AI uses when importing your models. See dbt's catalog docs: Catalog JSON file.

4. Sync dbt models to Wren AI​

wren dbt create

You will be prompted for the path to your dbt project and a project display name.

wren-create-project

This command will:

  • Locate and parse profiles.yml
  • Read catalog.json and manifest.json from the target/ directory
  • Build a Wren MDL (Model Definition Language), includes models, columns and relationships.
  • Create the project in Wren AI Cloud
  • Store the project configuration in the .wrenconfig file within your dbt project directory. This file contains the Wren AI Cloud Project ID. ⚠️ Ensure you commit this file to your dbt project repository, allowing other team members to execute wren dbt update to refresh the project.
Model selection behavior

Wren imports analytics-ready dbt models (e.g., models in the marts layer) and ignores staging/intermediate layers.

After the project is created, you could visit the project page to see the models and relationships. The relationships are extracted from the manifest.json file and the tests of the models.

wren-models

Additionally, you could check the connection details in the project settings.

wren-conneciton

5. Update an existing project​

Synchronize your linked Wren AI project (referenced in .wrenconfig) with the latest changes from your dbt models.

wren dbt update --dbt-path /path/to/dbt-project

Skip confirmation prompt​

wren dbt update --dbt-path /path/to/dbt-project --yes
warning

This command will replace the current MDL in your Wren AI project with the updated dbt models. Any manual changes made on the UI will be overwritten.