With Gemini Enterprise Agent Platform in express mode, you can sign up and begin building generative AI applications on Google Cloud. This simplified setup experience streamlines access to Google Cloud APIs by simplifying organization, billing, and project management.
To learn more about Agent Platform in express mode, see Google Cloud express mode FAQs.
Sign up for express mode
Agent Platform is available in express mode for developers with a
@gmail.com Google Account:
New users: New users to Google Cloud can sign up for an express mode account in a free tier to try Agent Platform for free for up to 90 days, within the specified quotas. You don't need to provide billing information to sign up in the free tier. You can upgrade to a paid tier by adding a billing account.
Agent Platform in express mode is separate from, and not available through, the Google Cloud Free Program. If you are in the Google Cloud Free Program, see the other quickstarts in the Get Started section.
To sign up for Agent Platform in express mode, open Express Mode from the Google Cloud console:
After you're set up, you can start using Agent Platform in express mode by following one of the tutorials:
Learn about Agent Platform in express mode
When you sign up for express mode, Google Cloud enables APIs on your behalf. This process happens in the background, so you can start using Agent Platform without needing to manually configure resources.
When you sign up for express mode, you get access to the following:
- Core Agent Studio features. Test and customize prompts for different generative AI models, and get the corresponding code to use in your application.
- An API key. Authenticate requests from your application.
- Free tier for new Google Cloud users during 90 days. During the 90 day limit of the free tier for new Google Cloud users, you can use the Agent Platform APIs that support express mode for free, up to their quotas. You don't have to provide billing information. You can increase your quota limits at any time by enabling billing.
If you are an existing Google Cloud user or after enabling billing, the 90 day free tier is removed. You transition to a paid tier to access extended quota limits and additional Google Cloud features. As your quotas are increased, you only pay for what you use. At any time, you can choose to end express mode and start using all the Google Cloud services and capabilities.
The following table lists the differences between the express mode experiences and the full Agent Platform experience:
| Item | Agent Platform express mode in the free tier | Full Agent Platform |
|---|---|---|
| Time limit | 90 days | Unlimited |
| Available services | Basic generative AI services. | All Google Cloud services, including Agent Platform. |
| Data sources | Google Drive | All data sources available in Google Cloud. |
| Quota | Free tier limits. See Available models and rate limits in express mode. | Standard pay-as-you-go limits. See Rate limits. |
| Standard format of API endpoints | Use the global endpoint without a project ID or location, and pass your API key in the x-goog-api-key HTTP header (or the key query parameter). For example:https://aiplatform.googleapis.com/v1/publishers/google/models/MODEL_ID:streamGenerateContent
|
Specify a project ID and location. For example:https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/publishers/google/models/MODEL_ID:streamGenerateContent
|
Available models and rate limits in express mode
You can try several models in express mode, including Gemini Flash models. The following table lists the models that are available in express mode, along with their rate limits:
| Model category | Available models | Maximum requests per minute |
|---|---|---|
| Gemini | gemini-3-pro-image |
Dynamic |
gemini-3.1-flash-image |
Dynamic | |
gemini-3.1-pro-preview |
Dynamic | |
gemini-3-flash-preview |
Dynamic | |
gemini-2.5-pro |
10 | |
gemini-2.5-flash-image |
10 | |
gemini-2.5-flash |
10 | |
gemini-2.5-flash-lite |
10 |
Only the models listed in the preceding table are supported in express mode.
Dedicated Gemini Live API models, such as
Gemini 3.8 Live
(gemini-3.8-live) and
Gemini 2.5 Flash Live API Native Audio
(gemini-live-2.5-flash-native-audio), aren't supported in
express mode, and Gemini Live API isn't available in the
Google Cloud console in express mode. To use
supported Gemini Live API models,
upgrade from express mode to a standard Google Cloud
account.
Agent Platform in express mode workflow
You can start sending requests from your application to Agent Platform APIs in three steps:
After signing up for express mode, use Agent Studio to try Agent Platform features:
For example, select Agent Studio > Create prompt to create and optimize multimodal prompts using a variety of Gemini models.
Get the code for what you implemented with the UI.
On the prompt page, click Build with code > Get code. A panel opens showing code that programmatically sends the same requests that you implemented in the UI. You can get the code for a programming language or curl. You can use Google Colab to try the Python code.
Use your API key to authenticate with the Agent Platform API.
In the Google Cloud console in express mode, open APIs & Services > Credentials:
Then, copy your Generative Language API Key into your code where it says
"YOUR_API_KEY". For example:Python
The Google Gen AI SDK for Python is available on PyPI and GitHub:
To learn more, see the Python SDK reference (opens in a new tab).
What's different in express mode
Agent Platform in express mode provides a subset of the features for generative AI. Therefore, some of the Agent Platform documentation is not relevant if you signed up in express mode. For details about the available API endpoints in express mode, see the Agent Platform in express mode REST API reference.
In addition, customers in Google Cloud typically use organizations and projects to work with resources (for example, to call an API endpoint). When using Agent Platform in express mode, you don't need to worry about organizations or projects. However, you might see them mentioned in some of the Google Cloud documentation that you reference while you're using Agent Platform in express mode. You can still use the documentation, but ignore concepts and instructions that refer to organizations and projects. In addition, the location you selected when signing up in express mode is used throughout your experience.
When calling REST API endpoints in express mode, use the global endpoint format for express mode and authenticate with your API key:
| Standard endpoint URL | https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/publishers/google/models/MODEL_ID:streamGenerateContent |
| Endpoint URL in express mode | https://aiplatform.googleapis.com/v1/publishers/google/models/MODEL_ID:streamGenerateContent |
Keep the following endpoint and authentication details in mind when sending REST requests in express mode:
- Publisher model path (
{model}): In the Agent Platform in express mode REST API reference, endpoint templates such ashttps://aiplatform.googleapis.com/v1/{model}:streamGenerateContentandhttps://aiplatform.googleapis.com/v1beta1/{model}:streamGenerateContentuse{model}to represent the full publisher model resource path (publishers/google/models/MODEL_ID). Both resolve to the same endpoint URL. - API key authentication header: Passing your API key as a query
parameter isn't required. Pass your API key in the
x-goog-api-keyHTTP header (for example,-H "x-goog-api-key: API_KEY"). If your client environment doesn't let you set HTTP headers, you can pass thekeysystem query parameter instead (?key=API_KEY). For more information, see Use API keys.
View and manage API keys
To authenticate with Agent Platform API endpoints that support express mode, use the API key that was created for you during sign-up or any key that you've created in express mode. An API key is an encrypted string that is auto-generated for you when you sign up in express mode. These API keys can be viewed and managed on the APIs & Services page.
To learn more about the best practices for managing API keys, see Best practices for managing API keys.
To view and manage your API keys, do the following:
In the Google Cloud console, go to the API Keys page:
Enable and manage billing
You can increase your quotas and remove the 90 day limit by enabling billing.
After enabling billing, you pay only for what you use. You can also save your prompts and use additional settings in the Google Cloud console that are grayed out when billing isn't enabled.
To manage billing, do the following:
Go to the Agent Studio Overview page in express mode.
In the Google Cloud console in express mode, open Billing:
Manage your billing accounts.
Start using all Google Cloud capabilities and services
Express mode is designed to help you get started. When you're ready to use other Google Cloud services or need more control over your environment, you can transition your express mode account to a standard Google Cloud account. This process is sometimes called upgrade.
You can start using all the capabilities and services available in Google Cloud in your project by upgrading your express mode account.
To upgrade from express mode, do the following:
Go to the Agent Studio Overview page in express mode.
Click Activate now and follow the instructions.
After you upgrade from express mode, specify your project ID and location instead of your API key when you call the REST API endpoints. For example:
https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/publishers/google/models/MODEL_ID:streamGenerateContent