---
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


---

# Set up a connection and credential alias for a third-party embedding model

# Set up a connection and credential alias for a third-party embedding model {#ariaid-title1}

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

Set up a connection and credential alias to authenticate an integration between your ServiceNow instance and a third-party embedding model.

## Before you begin

Role required: admin

## About this task

The system provides the default connection and credential alias for the Azure OpenAI Embedding model (text-embedding-3-large) and Google Gemini Embedding model (gemini-embedding-001). You set up these default aliases to manage the secure connection of your ServiceNow instance with either embedding model. A connection and credential alias includes the endpoint URL of the models and the login details, such as the OAuth token or API key. A connection and
credential alias tells the system how to connect to these third-party models.

## Procedure

1. Navigate to AllConnections \& CredentialsConnection \& Credential Aliases.
2. Select an alias that you want to set up.
   * To set up the embedding model alias for Azure OpenAI Embedding, select Azure OpenAI.
   * To set up the embedding model alias for Google Gemini Embedding, select Google Gemini API.
   {#setup-alias-for-3p-embedding-model__choices_wvw_pnd_1gc}
3. Create a connection record for your alias.
   1. In the Connections related list, select New.
   2. On the form, fill in the fields.  
      For a description of the field values, see [Create an HTTP(s) connection](https://www.servicenow.com/docs/access?context=create-https-connection&version=xanadu&pubname=xanadu-platform-security&ft:locale=en-US).
   3. Select Submit.

   {#setup-alias-for-3p-embedding-model__substeps_cnv_ftg_hgc}  
   The connection record is created.
4. Create a credential record for your alias.
   1. In the Connections related list, select a connection record from the connection alias list.  
      * Select the connection record that you created in step 3 for the Azure OpenAI alias.
      * Select the connection record that you created in step 3 for the Google Gemini API alias.
      {#setup-alias-for-3p-embedding-model__ul_fqq_svg_hgc}
   2. In the Credentials field, select the search icon (![Lookup using list.]()) and then select New.
   3. From the list of credentials, select a credential type.  
      * To create a credential record for the Azure OpenAI alias, select API Key Credentials.
      * To create a credential record for the Google Gemini API alias, select OAuth 2.0 Credentials.
      {#setup-alias-for-3p-embedding-model__ul_ftg_wvg_hgc}
   4. On the form, fill in the fields.  
      For a description of the field values, see [API key credentials](https://www.servicenow.com/docs/access?context=API-key-credential-form&version=xanadu&pubname=xanadu-platform-security&ft:locale=en-US) for the Azure OpenAI alias or [OAuth 2.0 credentials](https://www.servicenow.com/docs/access?context=oauth-2-credentials&version=xanadu&pubname=xanadu-platform-security&ft:locale=en-US) for the Google Gemini API alias.
   5. Select Submit.
   6. Select Update.

   {#setup-alias-for-3p-embedding-model__substeps_l2f_y5g_hgc}  
   The credential record is created.

## What to do next

Activate the embedding model to start using it. For more information, see [Activate a third-party embedding model](https://servicenow-prod.fluidtopics.net/8mg7shqp3fMNcdbo3QbLGQ "Activate your preferred embedding model, whether third-party or custom, so that the AI Search Retrieval Augmented Generation (RAG) application knows which model to use for generating embeddings.").

*[\>]: and then