Editor’s Note: This post is part of our Google I/O 2024 series sharing the latest Google Maps Platform news from our annual developer conference. To learn more about how to build experiences using Gemini model capabilities for the Places API, register for our session on May 16. 

For nearly two decades, developers have used our products to create innovative mapping experiences. Today, we’re helping developers drive even more innovation by bringing Gemini model capabilities to Google Maps Platform, starting with the Places API. In addition to using rich information about more than 250 million businesses and places–like ratings, reviews and business hours–developers will also be able to show helpful, generative AI summaries of places and geographic areas right in their own apps and websites. Updated frequently, these summaries use the Gemini model to analyze the place information and insights from Google Maps’ community of over 300 million contributors–helping you display the most up-to-date information possible. 

Another update to the Places API is AI-powered contextual search results. When users search for places in your products, you’ll now be able to display reviews and photos related to their search, to help them more easily compare places and make decisions. These features are now available to all developers as an Experimental release. Contextual search results are available globally, and place and area summaries are available in the U.S., with plans to expand to more countries over time. Try the demo to see how it works.

Place and area summaries in the Places API, using the Gemini model

With these new features, it’ll be faster and easier for people to find the information they’re looking for in your experiences. You’ll also get the benefits of deeper insights and summaries, saving you from writing custom descriptions of places yourself.

Help your users quickly understand a place with short and long place summaries

Let’s say you have a restaurant-booking app and want to help your users understand which restaurant is best for their group. When they search for restaurants in your app, they’ll be able to quickly see all the most important information within a summary, like the house specialty, happy hour deals, and the vibe of the place. This can help your users more easily decide where to go without needing to research each restaurant in detail. 

To help you tailor the summaries to fit your products, they’re available in two lengths–shorter summaries (on average around 100 characters) and longer summaries (on average around 400 characters). They’re available for many types of places, including restaurants, shops, supermarkets, parks, and movie theaters–with more to come in the future. Using the Gemini model, the short summaries are now updated more frequently and are available for far more places in the U.S. than ever before, compared to manually written summaries. The longer summaries provide even richer overviews, and can include more insight than was previously available in the Places API, such as must-order dishes, ambience, and service quality for restaurants.

Highlight things to do in a geographic area with new area summaries

Sometimes, it can also be helpful to describe what there is to do in an area around a place. With new area summaries using the Gemini model, users can now see overviews of what shopping, restaurants, and attractions are within walking distance from a place, so you can help users assess what to do when there. For example, an automaker can provide drivers with summaries of places close to their EV charging stations–like coffee shops, restaurants or stores–so they can choose which one to visit while their car is charging. 

Provide more relevant reviews and photos with contextual search results

Lastly, we’re now using AI to make it easier for your users to understand why they see particular search results by including more context. So if you have an app that allows users to explore local restaurants, your users can search for “dog friendly restaurants” to see a list of relevant dining spots, along with relevant reviews and excerpts from reviews, and photos of dogs at the restaurants. This information can help your users make more informed decisions and give them higher confidence in the results you’re providing. 

A closer look at area summaries content

Now that you have an overview of the new features, let’s take a deeper technical look at how one of the features works and how you’d use it in your projects. 

The places.AreaSummary object, found in the response, aims to give users a comprehensive overview of what an area has to offer. It categorizes places into relevant topics and provides brief descriptions of each place, allowing users to quickly grasp the variety of options available in the vicinity. This is particularly useful for exploring an unfamiliar area or deciding where to go based on specific interests (e.g., finding coffee shops or restaurants).

Let’s look at an area summary for an electric vehicle charging station. We’ll pay special attention to data returned about nearby coffee shops.

AreaSummary { 
content_blocks: [ 
{ 
topic: "overview"
...
}
{ 
topic: "coffee"  // The subject of the content block.
content: { 
text: "Acme Bread offers soups, salads, and sandwiches, along with coffee and baked goods. Happy Coffee is a well-known coffee shop serving signature roasts, light bites, and quick service. Perky Coffee specializes in coffee, tea, and breakfast items in a family-friendly and healthy environment."
}
references: { } // Place IDs of the Places used to construct the content block.
}
{ 
topic: "restaurant"
...
}
{ 
topic: "store"
...
}
]
}

The AreaSummary object contains an array of objects called content_blocks, where each object represents a specific topic or category within the area. It organizes information about the area by grouping similar places together. Each block provides:

• A topic: The category of the content, such as "overview," "coffee," "restaurant," or "store."

• Content: A description of the places within that category, often mentioning specific names and their offerings.

• Reference places: An array of place IDs, referencing the places mentioned in the content. This connects the summary back to specific Place details.

Check out the documentation for more details on how to add area summaries, place summaries, and contextual search results to your projects.

Get started

You can try the new Gemini model capabilities for the Places API and contextual search results at no cost during the Experimental phase. To get started and see how the features work, check out our demo. You can also submit feedback and feature requests through the public issue tracker. And to learn more about how to build geospatial experiences with the Gemini model capabilities for Places API, register for our I/O session. And don’t forget to tag Google Maps Platform on LinkedIn. We can’t wait to see what you build.