Quotas and overages

Your subscription entitles you to certain quotas across product features. If you have a pay-as-you-go subscription, there are no feature quota limits; you pay for what you use. For seat-based subscriptions, unlike other Google Cloud services, Gemini Enterprise feature quotas don't require manual increase requests; they're automatically adjusted based on the number of licenses or seats that you purchase.

Feature and edition-based quotas

Except for storage and data indexing, feature quotas are pooled across users in the same edition within a project and location. Storage and data indexing quota is pooled per project and location across all users, but it is not segmented by the edition—all editions share a single storage and data indexing pool. For all other feature quotas, users with licenses for a specific edition share a single combined limit per project and location.

Use case 1: Multiple editions in one project and location

If you have users with different editions in the same project and location, quota pools are maintained separately for each edition except for storage and data indexing:

  • Standard feature quotas: Quotas are pooled separately by edition in each project and location. For example, if a project and location has 100 Standard edition licenses and 50 Plus edition licenses, your Standard edition users share a pool of 16,000 Assistant queries per day (100 licenses × 160 queries), and your Plus edition users share a separate pool of 10,000 queries per day (50 licenses × 200 queries). These feature quota pools don't combine across editions.
  • Storage and data indexing: Quotas are pooled per project and location across all users, regardless of edition. For example, if a project and location has 100 Standard edition licenses (30 GiB per user) and 50 Plus edition licenses (75 GiB per user), there is a single combined storage pool of 6,750 GiB ((100 × 30 GiB) + (50 × 75 GiB)) shared by all users in that project and location across all editions.

Use case 2: Quota pooling within a single edition

Within an edition's quota pool, the following rules apply. For example, if you have 10 Standard edition licenses, your users share a total pooled quota of 1,600 queries per day (10 licenses × 160 queries):

  • The 10 Standard edition users share a combined pool of 1,600 queries for the day.
  • One power user can use 500 queries, leaving 1,100 queries for the remaining users in that edition.
  • Individual usage isn't capped at the per-user limit; all users in the edition draw from their edition's shared pool until it is depleted.

Quotas for Standard and Plus editions

For information about quotas across Gemini Enterprise Standard and Plus editions, see the following table:

Gemini Enterprise Standard Gemini Enterprise Plus
Quotas Pooled or unpooled Pooled across users within the edition per project and location Pooled across users within the edition per project and location
Product Storage + data indexing 30 GiB 75 GiB
Assistant 160 queries per day 200 queries per day
No-code agent building tools - create 1 agent per day 10 agents per day
No-Code agent building tools - use Based on underlying agent actions
Video generation 2 video generation requests per day 3 video generation requests per day
Image generation 5 images per day 10 images per day
General purpose agents made by Google Gemini Notebook - use Gemini Notebook Enterprise limits1 Gemini Notebook Enterprise limits1
Gemini Notebook - create & publish Gemini Notebook Enterprise limits1 Gemini Notebook Enterprise limits1
Deep Research 3 per day 10 per day
AI developer tools2 Antigravity $10 per user per month3 $15 per user per month3

1 For information about Gemini Notebook Enterprise usage limits, see the Gemini Notebook Enterprise documentation.

2 The AI developer tools feature is available for invoiced accounts only. For more information, see Cloud Billing account types.

3 Although quotas are expressed as a monthly per-user amount, they are enforced on a rolling seven-day basis as a shared pool. Your organization's total weekly quota is calculated by taking the per-user monthly value, dividing it by four, and multiplying it by your project's total purchased licensed seats. Any user in your edition can draw from this combined weekly pool. Unused quota doesn't roll over to the next week, and individual user caps don't apply unless manually configured.

Quotas for EDU and EDU Pro editions

For information about quotas across Gemini Enterprise EDU and EDU Pro editions, see the following table:

Gemini Enterprise EDU Gemini Enterprise EDU Pro
Quotas (Pooled or Unpooled) Pooled across users within the edition per project and location Pooled across users within the edition per project and location
Product Storage + data indexing 5 GiB 50 GiB
Assistant2 40 queries per day 200 queries per day
No-code agent building tools - Create 1 agent per day 10 agents per day
No-Code agent building tools - use Based on underlying agent actions
Video generation 1 video generation requests per day 3 video generation requests per day
Image generation 2 images per day 10 images per day
General purpose agents made by Google Gemini Notebook - use Gemini Notebook Enterprise limits1 Gemini Notebook Enterprise limits1
Gemini Notebook - create & publish Gemini Notebook Enterprise limits1 Gemini Notebook Enterprise limits1
Deep Research 1 per day 10 per day
Security and compliance Model Armor Same quota as the assistant Same quota as the assistant

1 For information about Gemini Notebook Enterprise usage limits, see the Gemini Notebook Enterprise documentation.

2 This includes the usage of Google Search and Web Grounding.

Quotas for Frontline editions

For information about quotas across Gemini Enterprise Frontline Starter and Worker, see the following table:

Gemini Enterprise Frontline Starter Gemini Enterprise Frontline Worker
Quotas (Pooled or Unpooled) Pooled across users within the edition per project and location Pooled across users within the edition per project and location
Product Storage + data indexing 2 GiB 2 GiB
Search Abuse preventative only2
Assistant 20 queries per day 40 queries per day
No-code agent building tools - create n/a n/a
No-Code agent building tools - use Based on underlying agent actions
Video generation n/a 1 video generation request per day
Image generation n/a 2 images per day
General-purpose agents made by Google Gemini Notebook - use Gemini Notebook Enterprise limits1 Gemini Notebook Enterprise limits1
Gemini Notebook - create & publish n/a n/a
Deep Research n/a 1 per day
Security and Compliance Model Armor Equivalent to assistant quota

1 For information about Gemini Notebook Enterprise usage limits, see the Gemini Notebook Enterprise documentation.

2Subject to abuse-preventative usage limits only.

Quotas for Emerging Market editions

For information about quotas across Gemini Enterprise Emerging Market editions, see the following table:

Gemini Enterprise Standard Emerging Market Gemini Enterprise Emerging Market EDU and governments
Quotas (Pooled or Unpooled) Pooled across users within the edition per project and location Pooled across users within the edition per project and location
Product Storage + data indexing 10 GiB 3 GiB
Search Abuse preventative only2 5 queries per day
Assistant 40 queries per day 5 queries per day, individual only
No-code agent building tools - create 1 agent per day 1 agent per day
No-Code agent building tools - use Based on underlying agent actions
Video generation n/a n/a
Image generation n/a 1 image per day
General-purpose agents made by Google Gemini Notebook Enterprise - use Gemini Notebook Enterprise limits1 n/a
Gemini Notebook - create & publish Gemini Notebook Enterprise limits1 n/a
Deep Research 1 per day n/a
Security and Compliance Model Armor Same quota as the assistant Same quota as the assistant

1 For information about Gemini Notebook Enterprise usage limits, see the Gemini Notebook Enterprise documentation.

2Subject to abuse-preventative usage limits only.

API and system quotas

This section lists the technical infrastructure quotas and system limits for Discovery Engine API resources used by Gemini Enterprise. System limits are fixed values that cannot be changed.

Allocation quotas

The following table lists the allocation quotas for Discovery Engine API. These quotas don't reset over time. Instead, they're released when you release the resource.

Google Cloud uses quotas to help ensure fairness and reduce spikes in resource use and availability. A quota restricts how much of a Google Cloud resource your Google Cloud project can use. Quotas apply to a range of resource types, including hardware, software, and network components. For example, quotas can restrict the number of API calls to a service, the number of load balancers used concurrently by your project, or the number of projects that you can create. Quotas protect the community of Google Cloud users by preventing the overloading of services. Quotas also help you to manage your own Google Cloud resources.

The Cloud Quotas system does the following:

In most cases, when you attempt to consume more of a resource than its quota allows, the system blocks access to the resource, and the task that you're trying to perform fails.

Quotas generally apply at the Google Cloud project level. Your use of a resource in one project doesn't affect your available quota in another project. Within a Google Cloud project, quotas are shared across all applications and IP addresses.

For more information, see the Cloud Quotas overview.

Quota Value
Total number of data stores per project 1001
Total number of engines per project 1502
Regional number of data stores per project per location (Global, US, EU) 100
Regional number of documents per project per location (Global, US, EU) 10,000,000
Regional number of engines per project per location (Global, US, EU) 150

1 Due to a technical limitation, the maximum quota for data stores is 500 per project. If you need more data stores, use new projects.
Each entity in a data store is counted as a data store instance and, therefore, counts toward this quota. For example, a data store with three entities counts as three data stores against your quota.

2 Due to a technical limitation, the maximum quota for engines is 500 per project. If you need more engines, use new projects.

Rate quotas

The following rate quotas apply to Discovery Engine API requests.

There are also system limits on Gemini Enterprise resources. System limits can't be changed.

Quota Value
Complete query requests per minute per project 300
Regional search requests per minute per project per location (Global, US, EU) 300

Overages

Feature usage is restricted when you exceed the feature quotas of your Gemini Enterprise edition. To avoid interruptions, a Gemini Enterprise administrator can turn on the Overages setting. Overage charges apply as defined in the following table for Gemini Enterprise Standard, Plus, and Emerging Markets editions. Other editions don't support overages.

If you have a Pay-as-you-go subscription, you pay for usage at the overage rates listed in the following table.

Prices are listed in US Dollars (USD). If you pay in a currency other than USD, the prices listed in your currency on Cloud Platform SKUs apply.

Gemini Enterprise
Quota units
Gemini Enterprise
Add-on Overage Charges
Quotas (Pooled or Unpooled) Pooled across users within the edition per project and location Pooled across users within the edition per project and location
Foundational engagement layer Storage + data indexing GiB $5 per GiB per month
Assistant Queries per day Agent Platform prices
No-code agent building tools - create Agents per day n/a
No-code agent building tools - use Based on underlying agent actions Agent Platform prices
Video generation Video generation requests per day Agent Platform prices
Image generation Images per day Agent Platform prices
General purpose agents made by Google Gemini Notebook Enterprise - use Gemini Notebook Enterprise limits1 n/a
Gemini Notebook Enterprise - create & publish Gemini Notebook Enterprise limits1 n/a
Deep Research Queries per day Agent Platform prices
AI developer tools2 Antigravity Dollars per month3 Agent Platform prices

1 For information about Gemini Notebook Enterprise usage limits, see the Gemini Notebook Enterprise documentation.

2 The AI developer tools feature is available for invoiced accounts only. For more information, see Cloud Billing account types.

3 Although quotas are expressed as a monthly per-user amount, they are enforced on a rolling seven-day basis as a shared pool. Your organization's total weekly quota is calculated by taking the per-user monthly value, dividing it by four, and multiplying it by your project's total purchased licensed seats. Any user in your edition can draw from this combined weekly pool. Unused quota doesn't roll over to the next week, and individual user caps don't apply unless manually configured.