Google is adding new features for third-party developers building on the Gemini AI model, which likely won’t be readily available from rivals like OpenAI’s ChatGPT and Anthropic’s Claude, as well as China’s growing number of open source options. It’s based on Google Maps.
This addition allows developers to connect the inference capabilities of Google’s Gemini AI model with live geospatial data from Google Maps, allowing applications to provide detailed, location-related responses to user queries, such as opening hours, reviews, and the vibe of a particular venue.
By leveraging data from over 250 million locations, developers can now build more intelligent and responsive location-aware experiences.
This is especially useful for applications where proximity, real-time availability, or location-specific personalization are important, such as local search, delivery services, real estate, and travel planning.
If the user’s location is known, developers can improve the quality of the response by passing latitude and longitude in the request.
By tightly integrating real-time and historical Maps data into the Gemini API, Google enables applications to generate well-founded, location-specific responses with factual accuracy and depth of context. This is only possible through the company’s mapping infrastructure.
Combining AI and geospatial intelligence
This new feature is accessible through Google AI Studio, where developers can try out a live demo using the Gemini Live API. The following models support grounding in Google Maps:
-
gemini 2.5 pro
-
gemini 2.5 flash
-
gemini 2.5 flashlight
-
gemini 2.0 flash
In one demonstration, a user asked for a recommended Italian restaurant in Chicago.
The assistant leveraged map data to find the top-rated options, uncover misspelled restaurant names, and then find the correct venue with accurate business details.
Developers can also obtain a context token to embed a Google Maps widget in their app’s user interface. This interactive component displays photos, reviews, and other familiar content typically found on Google Maps.
The integration is generateContent Gemini API methods. Developers include: googleMaps As a tool. You can also enable the map widget by setting a parameter in your request. Widgets rendered using the returned context token can provide a visual layer along with AI-generated text.
Industry-wide use cases
The Maps grounding tool is designed to support a variety of real-world use cases.
-
Generate itinerary: The travel app allows you to create detailed daily plans with route, timing, and venue information.
-
Personalized local recommendations: Real estate platforms can highlight properties near children’s facilities such as schools and parks.
-
Query detailed location: Applications can use community reviews and map metadata to provide specific information, such as whether a cafe has outdoor seating.
We recommend that developers enable tools only when geographic context is relevant to optimize both performance and cost.
According to the developer’s documentation, pricing starts at $25 per 1,000 fixed prompts, which is expensive for people who traffic large numbers of queries.
Combine search and maps to enhance context
Developers can mix Google Maps grounding and Google Search grounding in the same request.
Map tools provide factual data such as addresses, business hours, and ratings, while search tools add broader context from web content such as news and event listings.
For example, if you ask about live music on Beale Street, the integrated tool will provide venue details from a map and event times from a search.
According to Google, internal testing shows that using both tools together significantly improves response quality.
Customization and developer flexibility
The experience is built for customization. Developers can adjust system prompts, choose from a variety of Gemini models, and configure audio settings to tailor interactions.
Google AI Studio demo apps are remixable, allowing developers to test ideas, add functionality, and iterate designs within a flexible development environment.
The API returns structured metadata such as source links, location IDs, and citation ranges. Developers can use it to build inline citations and validate output generated by AI.
This supports transparency and increases trust in user-facing applications. Google also requires that map-based sources be clearly attributed and linked to the source using a URI.
AI Builder implementation considerations
For technical teams integrating this feature, Google recommends the following:
-
For better results, pass the user’s location context if known.
-
Show Google Maps source links directly below related content.
-
Enable the tool only when the query clearly includes geographic context.
-
Monitor latency and disable grounding when performance is important.
Grounding using Google Maps is currently available worldwide, but is prohibited in some regions (such as China, Iran, North Korea, and Cuba) and is not allowed for emergency response use cases.
Availability and access
Grounding with Google Maps is now generally available through the Gemini API.
With this release, Google continues to expand the capabilities of the Gemini API, enabling developers to build AI-driven applications that understand and respond to the world around them.
