1. Under the data loader block you just added, click the button dbt model, then click the option Single model.
  2. In the file browser that pops up, click the file of the dbt model you want to add to your pipeline.
  3. In the dbt profile target input field, enter the name of the dbt connection profile (e.g. dev) that you want to use when running the selected dbt models.

​
Depending on upstream blocks

If your dbt model references other models, those models will also be added to the current pipeline as upstream blocks. Once you’ve added 1 or more dbt models to your pipeline, you can set its dependencies on other blocks. The dbt model won’t run until all upstream blocks have successfully been completed.

​
Preview dbt model results

You can run a dbt model block and see the results of the SQL query. Under the hood, Mage is running dbt compile for that single model, then executing the compiled SQL query in the data source from your dbt project’s connection profile target.

​
Pipeline execution run

When pipeline is triggered and executes, it’ll run each dbt model block using the dbt run command.

​
Variable interpolation

You can add variables specific to your pipeline. These variables are accessible in each block of your pipeline. In addition, all the environment variables are accessible within the SQL query using the var syntax.
SyntaxDescriptionExample
{{ var('...') }}Get a value from the environment variables or runtime variables.{{ var('ds') }}

​
SQL statement

Within the SQL query in a dbt model block, you can access these variables using the var syntax. To learn more, read dbt’s documentation. For example: 
SELECT *
FROM users_v1
WHERE created_at >= '{{ var("ds") }}'
In the above example, the variable named ds is defined as part of the runtime variables for the pipeline.

​
Adding variables when running a YAML dbt block

In addition to interpolating variables described above, you can define variables directly in the dbt code block if it is a YAML dbt code block (i.e. not a SQL dbt block) by following these steps:
  1. In the Pipeline Editor, add a dbt model by selecting on the All models (w/option exclusion) option from the dropdown when clicking on the dbt model button for adding a block.
  2. For the newly added code block, you should see that is YAML type indicated in the top left corner of the block.
  3. On the same line as your --select or --exclude syntax, add the --vars syntax with the variables you want to include. Note: Putting the --vars syntax on a separate line may cause issues.
  4. Format the variables like a JSON object, so there should be double quotes around the keys and string values. Single quotes surrounding the variables object are optional.
  5. You can interpolate global or environment variables using double brackets around the variable name. See below for examples.
Examples of YAML dbt code block content (note the syntax for the variables object): 
--select demo/models --vars '{"demo_key": "demo_value", "date": 20230101}'
--select demo/models --vars {"demo_key":"demo_value","date":20230101}
--select demo/models --vars '{"global_var": {{ test_global_var }}, "env_var": {{ test_env_var }}}'
--select demo/models --vars {"refresh":{{page_refresh}},"env_var":{{env}}}