Yes, you can configure Elementary to only store test run results and avoid invoking the dbt-artifacts models, which can help with performance. To do this, you should set the following variables in your

dbt_project.yml

or via the CLI:

yaml vars:   disable_dbt_artifacts_autoupload: true   disable_dbt_invocation_autoupload: true   disable_run_results: false   disable_tests_results: false

-

disable_dbt_artifacts_autoupload: true

will prevent the metadata models (like

dbt_models

,

dbt_sources

, etc.) from being updated. -

disable_dbt_invocation_autoupload: true

will prevent invocation metadata from being uploaded. - Keeping

disable_run_results

and

disable_tests_results

as

false

ensures that test run results are still stored in your data warehouse. You can also set these variables conditionally based on your environment if needed. For example, to only enable artifact uploads in production:

yaml vars:   disable_dbt_artifacts_autoupload: "{{ target.name != 'prod' }}"   disable_dbt_invocation_autoupload: "{{ target.name != 'prod' }}"

This approach is commonly used to improve performance, especially in dbt Cloud or CI environments where only test results are needed and metadata uploads are unnecessary. Several users have reported significant performance improvements by disabling artifact uploads and only running them on deployment or as needed, rather than on every run or test execution [Disable on-run-end hooks](https://docs.elementary-data.com/dbt/on-run-end_hooks#can-i-disable-the-on-run-end-hooks?), [Slack thread: performance drift](https://slack.com/archives/C02CTC89LAX/p1717515678662739). If you want to be extra sure, you can also pass these variables directly in your dbt Cloud job or CLI command:

dbt run --vars '{"disable_dbt_artifacts_autoupload": true, "disable_dbt_invocation_autoupload": true}'

This will ensure that only test run results are stored, and the dbt-artifacts models are not invoked, minimizing performance overhead.