model_versions

Creates, updates, deletes, gets or lists a model_versions resource.

Overview

Name	model_versions
Type	Resource
Id	databricks_workspace.catalog.model_versions

Fields

The following fields are returned by SELECT queries:

get

Name	Datatype	Description
id	string	The unique identifier of the model version
metastore_id	string	The unique identifier of the metastore containing the model version
run_id	string	MLflow run ID used when creating the model version, if ``source`` was generated by an experiment run stored in an MLflow tracking server
run_workspace_id	integer	ID of the Databricks workspace containing the MLflow run that generated this model version, if applicable
catalog_name	string	The name of the catalog containing the model version
model_name	string	The name of the parent registered model of the model version, relative to parent schema
schema_name	string	The name of the schema containing the model version, relative to parent catalog
▶aliases	array
comment	string	The comment attached to the model version
created_at	integer
created_by	string	The identifier of the user who created the model version
▶model_version_dependencies	object	Model version dependencies, for feature-store packaged models
source	string	URI indicating the location of the source artifacts (files) for the model version
status	string	Current status of the model version. Newly created model versions start in PENDING_REGISTRATION status, then move to READY status once the model version files are uploaded and the model version is finalized. Only model versions in READY status can be loaded for inference or served. (FAILED_REGISTRATION, MODEL_VERSION_STATUS_UNKNOWN, PENDING_REGISTRATION, READY)
storage_location	string	The storage location on the cloud under which model version data files are stored
updated_at	integer
updated_by	string	The identifier of the user who updated the model version last time
version	integer	Integer model version number, used to reference the model version in API requests.

get_by_alias

Name	Datatype	Description
id	string	The unique identifier of the model version
metastore_id	string	The unique identifier of the metastore containing the model version
run_id	string	MLflow run ID used when creating the model version, if ``source`` was generated by an experiment run stored in an MLflow tracking server
run_workspace_id	integer	ID of the Databricks workspace containing the MLflow run that generated this model version, if applicable
catalog_name	string	The name of the catalog containing the model version
model_name	string	The name of the parent registered model of the model version, relative to parent schema
schema_name	string	The name of the schema containing the model version, relative to parent catalog
▶aliases	array
comment	string	The comment attached to the model version
created_at	integer
created_by	string	The identifier of the user who created the model version
▶model_version_dependencies	object	Model version dependencies, for feature-store packaged models
source	string	URI indicating the location of the source artifacts (files) for the model version
status	string	Current status of the model version. Newly created model versions start in PENDING_REGISTRATION status, then move to READY status once the model version files are uploaded and the model version is finalized. Only model versions in READY status can be loaded for inference or served. (FAILED_REGISTRATION, MODEL_VERSION_STATUS_UNKNOWN, PENDING_REGISTRATION, READY)
storage_location	string	The storage location on the cloud under which model version data files are stored
updated_at	integer
updated_by	string	The identifier of the user who updated the model version last time
version	integer	Integer model version number, used to reference the model version in API requests.

list

Name	Datatype	Description
id	string	The unique identifier of the model version
metastore_id	string	The unique identifier of the metastore containing the model version
run_id	string	MLflow run ID used when creating the model version, if ``source`` was generated by an experiment run stored in an MLflow tracking server
run_workspace_id	integer	ID of the Databricks workspace containing the MLflow run that generated this model version, if applicable
catalog_name	string	The name of the catalog containing the model version
model_name	string	The name of the parent registered model of the model version, relative to parent schema
schema_name	string	The name of the schema containing the model version, relative to parent catalog
▶aliases	array
comment	string	The comment attached to the model version
created_at	integer
created_by	string	The identifier of the user who created the model version
▶model_version_dependencies	object	Model version dependencies, for feature-store packaged models
source	string	URI indicating the location of the source artifacts (files) for the model version
status	string	Current status of the model version. Newly created model versions start in PENDING_REGISTRATION status, then move to READY status once the model version files are uploaded and the model version is finalized. Only model versions in READY status can be loaded for inference or served. (FAILED_REGISTRATION, MODEL_VERSION_STATUS_UNKNOWN, PENDING_REGISTRATION, READY)
storage_location	string	The storage location on the cloud under which model version data files are stored
updated_at	integer
updated_by	string	The identifier of the user who updated the model version last time
version	integer	Integer model version number, used to reference the model version in API requests.

Methods

The following methods are available for this resource:

Name	Accessible by	Required Params	Optional Params	Description
get	select	full_name, version, deployment_name	include_aliases, include_browse	Get a model version.
get_by_alias	select	full_name, alias, deployment_name	include_aliases	Get a model version by alias.
list	select	full_name, deployment_name	include_browse, max_results, page_token	List model versions. You can list model versions under a particular schema, or list all model versions
update	update	full_name, version, deployment_name		Updates the specified model version.
delete	delete	full_name, version, deployment_name		Deletes a model version from the specified registered model. Any aliases assigned to the model version

Parameters

Parameters can be passed in the WHERE clause of a query. Check the Methods section to see which parameters are required or optional for each operation.

Name	Datatype	Description
alias	string	The name of the alias
deployment_name	string	The Databricks Workspace Deployment Name (default: dbc-abcd0123-a1bc)
full_name	string	The three-level (fully qualified) name of the model version
version	integer	The integer version number of the model version
include_aliases	boolean	Whether to include aliases associated with the model version in the response
include_browse	boolean	Whether to include model versions in the response for which the principal can only access selective metadata for
max_results	integer	Maximum number of model versions to return. If not set, the page length is set to a server configured value (100, as of 1/3/2024). - when set to a value greater than 0, the page length is the minimum of this value and a server configured value(1000, as of 1/3/2024); - when set to 0, the page length is set to a server configured value (100, as of 1/3/2024) (recommended); - when set to a value less than 0, an invalid parameter error is returned;
page_token	string	Opaque pagination token to go to next page based on previous query.

SELECT examples

get

Get a model version.

SELECT
id,
metastore_id,
run_id,
run_workspace_id,
catalog_name,
model_name,
schema_name,
aliases,
comment,
created_at,
created_by,
model_version_dependencies,
source,
status,
storage_location,
updated_at,
updated_by,
version
FROM databricks_workspace.catalog.model_versions
WHERE full_name = '{{ full_name }}' -- required
AND version = '{{ version }}' -- required
AND deployment_name = '{{ deployment_name }}' -- required
AND include_aliases = '{{ include_aliases }}'
AND include_browse = '{{ include_browse }}'
;

get_by_alias

Get a model version by alias.

SELECT
id,
metastore_id,
run_id,
run_workspace_id,
catalog_name,
model_name,
schema_name,
aliases,
comment,
created_at,
created_by,
model_version_dependencies,
source,
status,
storage_location,
updated_at,
updated_by,
version
FROM databricks_workspace.catalog.model_versions
WHERE full_name = '{{ full_name }}' -- required
AND alias = '{{ alias }}' -- required
AND deployment_name = '{{ deployment_name }}' -- required
AND include_aliases = '{{ include_aliases }}'
;

list

List model versions. You can list model versions under a particular schema, or list all model versions

SELECT
id,
metastore_id,
run_id,
run_workspace_id,
catalog_name,
model_name,
schema_name,
aliases,
comment,
created_at,
created_by,
model_version_dependencies,
source,
status,
storage_location,
updated_at,
updated_by,
version
FROM databricks_workspace.catalog.model_versions
WHERE full_name = '{{ full_name }}' -- required
AND deployment_name = '{{ deployment_name }}' -- required
AND include_browse = '{{ include_browse }}'
AND max_results = '{{ max_results }}'
AND page_token = '{{ page_token }}'
;

UPDATE examples

update

Updates the specified model version.

UPDATE databricks_workspace.catalog.model_versions
SET 
aliases = '{{ aliases }}',
catalog_name = '{{ catalog_name }}',
comment = '{{ comment }}',
created_at = {{ created_at }},
created_by = '{{ created_by }}',
id = '{{ id }}',
metastore_id = '{{ metastore_id }}',
model_name = '{{ model_name }}',
model_version_dependencies = '{{ model_version_dependencies }}',
run_id = '{{ run_id }}',
run_workspace_id = {{ run_workspace_id }},
schema_name = '{{ schema_name }}',
source = '{{ source }}',
status = '{{ status }}',
storage_location = '{{ storage_location }}',
updated_at = {{ updated_at }},
updated_by = '{{ updated_by }}'
WHERE 
full_name = '{{ full_name }}' --required
AND version = '{{ version }}' --required
AND deployment_name = '{{ deployment_name }}' --required
RETURNING
id,
metastore_id,
run_id,
run_workspace_id,
catalog_name,
model_name,
schema_name,
aliases,
comment,
created_at,
created_by,
model_version_dependencies,
source,
status,
storage_location,
updated_at,
updated_by,
version;

DELETE examples

delete

Deletes a model version from the specified registered model. Any aliases assigned to the model version

DELETE FROM databricks_workspace.catalog.model_versions
WHERE full_name = '{{ full_name }}' --required
AND version = '{{ version }}' --required
AND deployment_name = '{{ deployment_name }}' --required
;