---
sourceDocument: Xanadu ServiceNow AI Platform Administration
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/xanadu/platform-administration

 Release :

    - xanadu

ft:locale :

    - en-US

ft:publication_title :

    - Xanadu ServiceNow AI Platform Administration

ft:clusterId :

    - platadm

bundleId :

    - platadm

workflow :

    - Platform


---

# Enable the custom embedding model for semantic indexing

# Enable the custom embedding model for semantic indexing {#ariaid-title1}

* Release version: Xanadu
* 
* Updated July 31, 2025
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read

Add a new embedding model in the semantic indexing table so that the AI Search
Retrieval Augmented Generation (RAG) application can use this model for semantic indexing.

## Before you begin

Role required: admin

## Procedure

1. Navigate to All, and then enter <kbd class="ph userinput">ais_semantic_embedding_model.list</kbd> in the filter to go to the AI Search Semantic Embedding Models \[sys_generative_ai_config\] table.
2. Select New.
3. In the Name field, enter a unique name.  
   For example, Azure OpenAI Large Text Embedding.
4. In the Model Id field, enter a unique ID.  
   An ID starts with a letter or number and may include letters, digits, periods (.), or hyphens (-) after the first character.
5. In the One Extend Capability Definition field, select a BYOM capability definition that you created to set a provider for the embedding model.
6. In the Model Config field, select an embedding model that is already configured.
7. Select Active.
8. **Optional:** If you want to configure batching for your embedding model, do these steps:  
   Batching helps the embedding model to process multiple inputs at once. The minimum and maximum batch size values control how inputs are grouped and processed to call the embedding generation API.
   1. Select Batching Supported.
   2. In the Minimum Batch Size and Maximum Batch Size fields, enter the required values.  
      For example, the minimum number of inputs allowed in a single batch is 4 and the maximum number of inputs that can be processed together in one batch is 16.
   {#enable-byom-for-semnatic-indexing__substeps_a54_lmd_lgc}
9. In the Error Handler Extension Instance field, select an error handler instance.  
   You create a scripted extension point to handle embedding generation errors that occur when custom embedding models generate semantic vectors. For more information, see [Create an error handler extension point](https://servicenow-prod.fluidtopics.net/A29LXRaVRd1TJZ7nTmgknA "Create a scripted extension point to handle the embedding generation errors that occur when custom embedding models in the AI Search Retrieval Augmented Generation (RAG) application generate semantic vectors.").
10. Select Submit.

## What to do next

Add your embedding model to the semantic index configuration to enable content ingestion with that model. For more information, see [Configure semantic indexing settings for an indexed source](https://servicenow-prod.fluidtopics.net/MTroW~0rScnMYCEtTi4DlA "Specify the semantic indexing configuration settings you want to apply when AI Search indexes records from your indexed sources.").