Skip to main content

Materialize setup

Vendor-supported plugin

Certain core functionality may vary. If you would like to report a bug, request a feature, or contribute, you can check out the linked repository and open an issue.

Overview of dbt-materialize

Installing dbt-materialize

pip is the easiest way to install the adapter:

pip install dbt-materialize

Installing dbt-materialize will also install dbt-core and any other dependencies.

Configuring dbt-materialize

For Materialize-specifc configuration, please refer to Materialize Configuration.

For further info, refer to the GitHub repository: MaterializeInc/materialize/blob/main/misc/dbt-materialize

Connecting to Materialize

Once you have set up a Materialize account, adapt your profiles.yml to connect to your instance using the following reference profile configuration:

target: dev
type: materialize
host: [host]
port: [port]
user: []
pass: [password]
dbname: [database]
cluster: [cluster] # default 'default'
schema: [dbt schema]
sslmode: require
keepalives_idle: 0 # default: 0, indicating the system default
connect_timeout: 10 # default: 10 seconds
retries: 1 # default: 1, retry on error/timeout when opening connections


cluster: The default cluster is used to maintain materialized views or indexes. A default cluster is pre-installed in every environment, but we recommend creating dedicated clusters to isolate the workloads in your dbt project (for example, staging and data_mart).

keepalives_idle: The number of seconds before sending a ping to keep the Materialize connection active. If you are encountering SSL SYSCALL error: EOF detected, you may want to lower the keepalives_idle value to prevent the database from closing its connection.

To test the connection to Materialize, run:

dbt debug

If the output reads "All checks passed!", you’re good to go! Check the dbt and Materialize guide to learn more and get started.

Supported Features


Because Materialize is optimized for transformations on streaming data and the core of dbt is built around batch, the dbt-materialize adapter implements a few custom materialization types:

sourceYESCreates a source.
viewYESCreates a view.
materializedviewYESCreates a materialized view.
tableYESCreates a materialized view. (Actual table support pending #5266)
sinkYESCreates a sink.
ephemeralYESExecutes queries using CTEs.
incrementalNOUse the materializedview materialization instead. Materialized views will always return up-to-date results without manual or configured refreshes. For more information, check out Materialize documentation.


Materialized views (materializedview), views (view) and sources (source) may have a list of indexes defined.


Running dbt seed will create a static materialized view from a CSV file. You will not be able to add to or update this view after it has been created.


Running dbt test with the optional --store-failures flag or store_failures config will create a materialized view for each configured test that can keep track of failures over time.