This page describes serving configs for media recommendation apps and how tocreate, edit, and delete them. A serving config contains configurationsthat are used when generating your media recommendations results.

About serving configs

Serving configs are referenced at serving time to determine what recommendationsto generate.

When any Vertex AI Search app is created, a default serving config isalso automatically created.For recommendations apps, the serving config's ID is the sameas the engine ID.For search apps, the default serving config's ID isdefault_search.The default serving config ID is specified when makingrecommend orsearch API calls.

However, search and custom recommendations apps don't allow you tomodify serving configs or create additional serving configs. Only mediarecommendation apps have modifiable serving configs and allow multiple servingconfigs.

For any media recommendations app, you can create multiple serving configs withdifferent settings for recommendations boost/bury, demotion, and diversity.To get recommendations from a specific serving config with the boost/bury,demotion, and diversity settings that you need, specify that serving config's IDin therecommend request.

Serving config settings

The following settings are available for serving configs:

• Recommendation boost/bury. You can boost or bury media recommendations based ona filter and boost value. For information about boosting and burying mediarecommendations,Boost and bury media recommendations.

  Note: Boost/bury controls can only be created through the API. They can't becreated using Google Cloud console.

• Recommendation demotion. You can demote media recommendations based onviewer data or content age. For information about demotion settings and how toset them, seeDemote media recommendations.

• Recommendation diversity. You can diversify recommendations to reduce thelikelihood that similar content is shown in a recommendation panel. Forinformation about diversification settings and how to set them, seeDiversify media recommendations.

Create a serving config

To create a serving config, follow these steps:

1. In the Google Cloud console, go to theAI Applications page.

   AI Applications

2. Click the name of the app for which you will create a serving config.

3. Go to theConfigurations >Control tab.

4. ClickCreate new serving config.

5. Enter a name for your serving config.

6. Set configurations for recommendations: demotion and diversity.

   For more information about these settings, seeDemote mediarecommendations, andDiversify mediarecommendations.

7. ClickSave and publish.

Edit a serving config

To edit a serving config, follow these steps:

1. In the Google Cloud console, go to theAI Applications page.

   AI Applications

2. Click the name of the app whose serving config you want to edit.

3. Go to theConfigurations >Control tab.

4. If you have more than one serving config, click the name of the servingconfig that you want to edit.

5. Update your configurations for recommendation demotion ordiversity.

   For more information about these settings, seeDemote media recommendations andDiversify media recommendations.

6. ClickSave and publish.

Delete a serving config

You can delete a serving config only if you have multiple serving configs.

To delete a serving config, follow these steps:

1. In the Google Cloud console, go to theAI Applications page.

   AI Applications

2. Click the name of the app whose serving config you will delete.

3. Go to theConfigurations >Control tab.

4. For the serving config you that want to delete, clickmore_vert, and then clickDelete.

5. In the confirmation dialog, type the name of the serving config to delete,and then clickConfirm.