AI in Yandex Maps: Recommendations for Leisure Spots

The AI in Yandex Maps has learned to give personalized recommendations for leisure and cultural outings based on the user's preferences: opening hours, menu and type of activity.

What the AI assistant can do now

Previously, the AI assistant in Yandex Maps mainly handled navigation tasks and basic search by category. After the update, it acts as a conversational leisure advisor and accepts multi-part requests in natural language.

Users can describe a task in free form, such as "I want to have dinner at a restaurant with live fish that's open until midnight" or "find a skating rink with skate rental for kids on Saturday morning". The assistant processes all the conditions itself and suggests suitable options on the map. If the first results don't fit, the request can be refined in the conversation without starting the search over.

Criteria can be added step by step: start with "restaurants with live fish", get a list, then ask for "only those open until 23:00", and the assistant narrows the selection. Location refinements work the same way: "closer to the metro" or "in this area".

Key features:

• Opening hours — finds venues open on the day and hour you need
• Menu contents — specific dishes, drinks or cuisine at a restaurant or cafe
• Type of activity — sports and fitness, cultural outings, entertainment, walks, family leisure
• Conversational refinement — the request is adjusted over several steps without restarting the search

The assistant covers both venues (restaurants, cafes, bars) and activities (sports facilities, parks, cultural spaces): everything in the Yandex Maps database.

How this differs from search with filters

Standard search in mapping services is linear: the user picks a category, sets filters by hand and scrolls through a list. Every new criterion is an extra step. This works for simple queries but gets awkward when there are several conditions or they go beyond the standard filters.

A conversational AI assistant works differently: it takes a request in natural language and interprets it as a whole. Users don't have to work out which filter to use for "a coffee shop with a quiet atmosphere, power outlets and good wifi, open today after 20:00". The assistant treats such a request as a single unit and chooses the criteria itself.

The conversational format also lowers the barrier for less experienced users: it is enough to describe the situation in words ("there are three of us adults, we're traveling with kids and want something active on Saturday morning"), and the system does the rest. The familiar filters are not going away, but users now have an alternative. This brings interaction with maps closer to how people usually ask friends for advice.

Why such features are appearing now

Yandex Maps is one of the most visited mapping services in Russia and the CIS, with tens of millions of active users. Adding an AI leisure advisor fits a broader strategy among tech companies: embedding generative AI not as a standalone product but as a functional layer inside services people already use.

This approach puts an AI feature in front of the largest possible audience at once, without having to persuade users to download a new app. People use Yandex Maps as before, but now get a smart assistant inside the familiar interface. For Yandex it is also a way to increase engagement: a user who gets useful leisure tips returns to the app more often.

What this means

The Yandex Maps update is a small but telling step: the AI assistant stops being a separate chatbot and becomes part of the infrastructure of everyday life. For the end user, it means fewer steps between "I want to go somewhere" and "here is a specific address and a route to the place".

What new features were added to Yandex Maps in July 2026?

The built-in neural network now recommends leisure spots based on opening hours, menu contents and type of activity, using your preferences expressed in natural language.

How do I choose a place to relax using the neural network in

Yandex Maps? Describe what you need in free form, for example where to have dinner or spend a day off. The neural network will understand and pick suitable places.

What requests does the AI in Yandex Maps accept now?

Multi-part requests in natural language: for example, a user can describe a task in free form, and the AI will suggest suitable places based on opening hours, menu contents and type of activity.