Delete a model from Vertex AI Model Registry

Learn how to delete a model you no longer need from the Vertex AI Model Registry.

If you want to delete a BigQuery ML from Vertex AI Model Registry, you must first delete it from BigQuery ML. To learn more, see BigQuery ML and Vertex AI Model Registry.

If you want to delete a model that is currently deployed to an endpoint, you need to undeploy it first. Otherwise, you are unable to delete the model.

Delete a model

Console

Go to the Model Registry page from the Vertex AI section in the Google Cloud console.

Go to the Model Registry page
Select More actions from the model you want to delete.
Select Delete model. When you delete the model, all associated model versions and evaluations are deleted from your Google Cloud project.
Click Delete on the confirmation screen.

API

Delete a model using the Vertex AI SDK for Python.

Python


from google.cloud import aiplatform


def delete_model_sample(model_id: str, project: str, location: str):
    """
    Delete a Model resource.
    Args:
        model_id: The ID of the model to delete. Parent resource name of the model is also accepted.
        project: The project.
        location: The region name.
    Returns
        None.
    """
    # Initialize the client.
    aiplatform.init(project=project, location=location)

    # Get the model with the ID 'model_id'. The parent_name of Model resource can be also
    # 'projects/<your-project-id>/locations/<your-region>/models/<your-model-id>'
    model = aiplatform.Model(model_name=model_id)

    # Delete the model.
    model.delete()

Delete a model version

Console

Go to the Model Registry page from the Vertex AI section in the Google Cloud console.

Go to the Model Registry page
Expand the model to view its model versions. Select the version that you want to delete.
Select the More actions from the model version. menu .
Select Delete version. All associated model evaluations are deleted when you delete your version.

API

Python


from google.cloud import aiplatform


def delete_model_version_sample(
    model_id: str, version_id: str, project: str, location: str
):
    """
    Delete a Model version.
    Args:
        model_id: The ID of the model to delete. Parent resource name of the model is also accepted.
        version_id: The version ID or version alias of the model to delete.
        project: The project ID.
        location: The region name.
    Returns
        None.
    """
    # Initialize the client.
    aiplatform.init(project=project, location=location)

    # Initialize the Model Registry resource with the ID 'model_id'.The parent_name of Model resource can be also
    # 'projects/<your-project-id>/locations/<your-region>/models/<your-model-id>'
    model_registry = aiplatform.models.ModelRegistry(model=model_id)

    # Delete the model version with the version 'version'.
    model_registry.delete_version(version=version_id)

Delete a model version with the default alias

Console

From the Model Registry, select the model name to view its model versions.
Select the version that you want, and from the Actions button click Delete. A warning opens since you're attempting to delete the default alias version. Set another version as default first.
Select what version you want to make default for the model from the dropdown.
On the confirmation screen, click Set and delete.

API

Python


from typing import List

from google.cloud import aiplatform


def delete_aliases_model_version_sample(
    model_id: str,
    version_aliases: List[str],
    version_id: str,
    project: str,
    location: str,
):
    """
    Delete aliases to a model version.
    Args:
        model_id: The ID of the model.
        version_aliases: The version aliases to assign.
        version_id: The version ID of the model to assign the aliases to.
        project: The project ID.
        location: The region name.
    Returns
        None.
    """
    # Initialize the client.
    aiplatform.init(project=project, location=location)

    # Initialize the Model Registry resource with the ID 'model_id'.The parent_name of Model resource can be also
    # 'projects/<your-project-id>/locations/<your-region>/models/<your-model-id>'
    model_registry = aiplatform.models.ModelRegistry(model=model_id)

    # Remove the version aliases to the model version with the version 'version'.
    model_registry.remove_version_aliases(
        target_aliases=version_aliases, version=version_id
    )