---
title: Model Registry workflow update
description: Between the October and November 2023 AI Platform releases (and in the 9.2 Self-managed AI Platform release), DataRobot is launching an exciting update to our Model Registry, making it easier for you to organize your models and track multiple versions.

---

# Model Registry workflow update

Between the October and November 2023 AI Platform releases (and in the 9.2 Self-managed AI Platform release), DataRobot is launching an exciting update to our Model Registry, making it easier for you to organize your models and track multiple versions. No action is required from you; however, once this change is rolled out, you must register a model prior to deployment.

## Migration and workflow update overview

The new Model Registry is an organizational hub for the variety of models used in DataRobot. Models are registered as deployment-ready model packages. These model packages are grouped into _registered models_ containing _registered model versions_, allowing you to categorize them based on the business problem they solve. Registered models can contain DataRobot, custom, external, challenger, and automatically retrained models as versions. 

During this update, packages from the **Model Registry > Model Packages** tab are converted to registered models and migrated to the new **Registered Models** tab. Each migrated registered model contains a registered model version. The original packages can be identified in the new tab by the model package ID appended to the registered model name:

![](images/migrate-model-package.png)

Once the migration is complete, in the updated **Model Registry**, you can track the evolution of your predictive and generative models with new versioning functionality and centralized management. In addition, you can access both the original model and any associated deployments and share your registered models (and the versions they contain) with other users:

![](images/rn-reg-models-page.png)

This update builds on the [previous model package workflow changes](release/index#model-package-artifact-creation-workflow), requiring the registration of any model you intend to deploy. To register and deploy a model from the Leaderboard, you must first provide model registration details:

* _Previously_, when you opened a model on the **Leaderboard** and navigated to the **Predict > Deploy**, you could click **Deploy model** without providing registered model details.

* _With this update_, when you open a model on the **Leaderboard** and navigate to the **Predict > Deploy** tab, you are prompted to **Register to deploy**, providing model details and adding the model to the Model Registry as a new registered model, or as a new version of an existing model. Once the model is registered, you can click **Deploy**.

![](images/register-before-deployment.png)

!!! tip "Deploy a model version directly from the Leaderboard"
    _If you have already registered the model_, on the **Leaderboard**, you can open the model's **Predict > Deploy** tab, locate the model in the **Model Versions** list, and click **Deploy**&mdash;even if the **Status** is **Building**.

## Leaderboard deployment walkthrough

To make the Model Registry a true organizational hub for all models in DataRobot, each model must be registered and _then_ deployed. To register and deploy a model from the Leaderboard:  

1. On the **Leaderboard**, select the model to use for generating predictions. DataRobot recommends a model with the **Recommended for Deployment** and **Prepared for Deployment** badges. The [model preparation](model-rec-process) process runs feature impact, retrains the model on a reduced feature list, and trains on a higher sample size, followed by the entire sample (latest data for date/time partitioned projects).

    ![](images/rn-prepared-for-deployment.png)

2. Click **Predict > Deploy**. If the Leaderboard model doesn't have the **Prepare for Deployment** badge, DataRobot recommends you click **Prepare for Deployment** to run the [model preparation](model-rec-process#prepare-a-model-for-deployment) process for that model.

    ![](images/rn-prepare-for-deployment-process.png)

    !!! tip
        If you've already added the model to the Model Registry, the registered model version appears in the **Model Versions** list. You can click **Deploy** next to the model and skip the rest of this process.

3. Under **Deploy model**, click **Register to deploy**.

    ![](images/rn-reg-dr-model.png)

4. In the **Register new model** dialog box, provide the required model package model information:

    ![](images/rn-reg-model-fields.png)

    | Field | Description |
    |-------|-------------|
    | Register model | Select one of the following:<ul><li>**Register new model:** Create a new registered model. This creates the first version (**V1**).</li><li>**Save as a new version to existing model:** Create a version of an existing registered model. This increments the version number and adds a new version to the registered model.</li></ul> |
    | Registered model name / Registered Model | Do one of the following:<ul><li>**Registered model name:** Enter a unique and descriptive name for the new registered model. If you choose a name that exists anywhere within your organization, the **Model registration failed** warning appears.</li><li>**Registered Model:** Select the existing registered model you want to add a new version to.</li></ul> |
    | Registered model version | Assigned automatically. This displays the expected version number of the version (e.g., V1, V2, V3) you create. This is always **V1** when you select **Register a new model**. |
    | Prediction threshold | _For binary classification models_. Enter the value a prediction score must exceed to be assigned to the positive class. The default value is `0.5`. For more information, see [Prediction thresholds](#prediction-thresholds). |
    | **Optional settings**  | :~~: | :~~: |
    | Version description | Describe the business problem this model package solves, or, more generally, describe the model represented by this version. |
    | Tags | Click **+ Add item** and enter a **Key** and a **Value** for each key-value pair you want to tag the model _version_ with. Tags do not apply to the registered model, just the versions within. Tags added when registering a new model are applied to **V1**. |
    | Include prediction intervals | _For time series models_, if you enabled the [time series model package prediction intervals](pp-ts-pred-intervals-mlpkg) feature, you can enable the computation of prediction intervals during the time series model package build process. For more information, see [Prediction intervals](#prediction-intervals). |

    ??? note "Binary classification prediction thresholds"
        If you set the <span id="prediction-thresholds">prediction threshold</span> before the [deployment preparation process](model-rec-process), the value does not persist. When deploying the prepared model, if you want it to use a value other than the default, set the value after the model has the **Prepared for Deployment** badge.

    
    ??? note "Public Preview: Time series prediction intervals"
        If the [time series model package prediction intervals](pp-ts-pred-intervals-mlpkg) feature is enabled, you can access the <span id="prediction-intervals">**Include prediction intervals**</span> setting when you register and deploy a time series model. When you deploy a model package with prediction intervals, the **Predictions > Prediction Intervals** tab is available in the deployment. For deployed model packages built without computing intervals, the deployment's **Predictions > Prediction Intervals** tab is hidden; however, older time series deployments without computed prediction intervals may display the **Prediction Intervals** tab if they were deployed prior to August 2022.

5. Click **Add to registry**. The model opens on the **Model Registry > Registered Models** tab.

6. While the registered model builds, click **Deploy** and then [configure the deployment settings](add-deploy-info).

    ![](images/rn-model-artifact-creation-building.png)

## API route deprecations

We are deprecating certain API routes to support this change. All APIs should function as expected for 6 months. API users can check out our [API documentation](api/index) for more details.

## Frequently asked questions

Question          | Answer
------------------|---------
What is changing? | The **Registered Models** tab is replacing the **Model Packages** tab. When you add a model to the **Model Registry**, it has a version number (`v1`, `v2`, etc.) and is called a _registered model_. Each _registered model_ contains _registered model versions_. In addition, the workflow to deploy Leaderboard models requires a model registration step.
How do I deploy Leaderboard models? | You must add a model to the Model Registry before deploying.<ol><li>On the Leaderboard, open a model and navigate to **Predict > Deploy**.</li><li>_New_: Click **Register to deploy** and register the model. You can register a new model (e.g., `v1`) or save as a new version of an existing registered model.</li><li>After registration, you are directed to the version in the **Model Registry**, where you can click **Deploy**.</li></ol>
Why are we making this change? | <ul><li>**Improved User Experience**: Requiring the registration of all deployed models improves the user experience for organizing models. If you retrain models, you can see the full lineage of that model. In addition, you can manually group models, filtering and searching for models is easier, and the workflow for challenger models is simpler.</li><li>**AI Production Functionality**: The new Model Registry is a centralized organizational hub for all models, regardless of where they are built or hosted. Registering models before deployment enables critical AI Production functionality, like monitoring and governance.
What happens to model packages created before the migration? | All existing model packages in the Model Registry are migrated as unique registered models containing a version. The original packages can be identified in the new tab by the model package ID appended to the registered model name.
