Make the MongoDB docs better! We value your opinion. Share your feedback for a chance to win $100.
Click here >
Docs Menu
Docs Home
/ /

Manage Automated Embedding

You can manage Automated Embedding from the Atlas UI.

To enable Automated Embedding, you must create a MongoDB Vector Search index with the autoEmbed type. To learn more, see Create an Automated Embedding Index.

To disable Automated Embedding at the organization level, use an Atlas Resource Policy. You can disable Automated Embedding entirely or disable it with project-level exceptions.

To disable Automated Embedding entirely, use the following syntax in your Atlas Resource Policy:

forbid (
principal,
action == ResourcePolicy::Action::"search.index.modify",
resource
)
when {
context.search.index.isAutoEmbed
};

To disable Automated Embedding with exceptions for specific projects, use the following syntax in your Atlas Resource Policy. Replace each <project-id> with the ID of the project to exempt. To find your project ID, see: Manage Project Settings.

forbid (
principal,
action == ResourcePolicy::Action::"search.index.modify",
resource
)
when {
context.search.index.isAutoEmbed
}
unless {
resource in ResourcePolicy::Project::"<project-id-1>" ||
resource in ResourcePolicy::Project::"<project-id-2>"
};

When active, these policies prevent new autoEmbed type indexes from being created. Indexes created before the resource policy was activated are not affected and must be deleted manually to maintain compliance.

To apply this policy to your organization:

1
  1. If it's not already displayed, select your desired organization from the Organizations menu in the navigation bar.

  2. In the sidebar, click Organization Settings.

2

In the sidebar, click Resource Policies under the Configurations header.

3
4
  1. Give the policy a unique name.

  2. (Optional) Add a description.

  3. In the Cedar Policy field, paste the Cedar policy. Replace <project-id> with your project ID.

5

To enable Automated Embedding, you must do the following:

To view aggregated token usage and embedding model requests for Automated Embedding service:

1
2

You can go the MongoDB Search page from the Search & Vector Search option, or the Data Explorer.

3
4

The Automated Embedding Usage page displays. The Automated Embedding Usage page displays the following information:

Section
Description

Usage Summary

Overview of your Automated Embedding usage, including total tokens used and remaining free tokens.

Usage by Model

Breakdown of token usage by embedding model.

Usage by Operation

Breakdown of token usage by operation type (indexing and querying).

5
1
2
  1. If it's not already displayed, select your desired organization from the Organizations menu in the navigation bar.

  2. At the organization level, click Usage under the AI Models header in the navigation bar.

3
4

The Usage page displays usage metrics for all model API keys in your organization. The page provides the following tabs:

  • Usage Activity: Shows detailed usage metrics and charts for your organization.

  • Free Usage: Shows free quota consumption by model.

To learn more, see Usage Page Features.

To view rate limits for Automated Embedding:

1
2
  1. If it's not already displayed, select your desired organization from the Organizations menu in the navigation bar.

  2. If it's not already displayed, select your desired project from the Projects menu in the navigation bar.

  3. In the sidebar, click Search & Vector Search under the Database header.

    The Search & Vector Search page displays.

3
4

The page displays the following information:

Name
Description

Model

List of Voyage AI embedding models.

Tokens Per Minute (TPM)

Number of tokens that you can process per minute.

Requests Per Min (RPM)

Number of API requests that you can send per minute.

1
2
  1. If it's not already displayed, select your desired organization from the Organizations menu in the navigation bar.

  2. At the organization level, click Rate Limits under the AI Models header in the navigation bar.

3

The page displays the following information:

Name
Description

Model

List of Voyage AI embedding models.

Tokens Per Minute (TPM)

Number of tokens that you can process per minute.

Requests Per Min (RPM)

Number of API requests that you can send per minute.

If you are on the Free Tier and need to increase rate limits for Automated Embedding, add a payment method to upgrade to a paid tier. To learn more, see Set Payment Method.

If you are on a Paid Tier, MongoDB automatically increases rate limits as your usage increases. If you still need a higher rate limit, contact MongoDB Support.

Back

Model Pricing and Rate Limits

On this page