Tencent Maps Unveils Full Agent Capability Map and Launches LBSwork Workbench

Tencent Maps unveils full Agent capability landscape! Yuanbao, Xiaowei, Rokid and others have all integrated it. Kuaikeji Download the app for exclusive early access. October 31, 2026, 10:47:08, from Henan. Updated October 31, 2026, 12:29:02

Kuaikeji News, October 31: At the "Tencent Maps Industry Application Session" of the Tencent Global Digital Ecosystem Summit on October 30, Tencent Maps officially unveiled its full Agent capability landscape to the public.

Also released at the same time was the desktop AI map workbench Tencent Maps LBSwork, which is based on the same-origin Harness framework as WorkBuddy and provides one-stop "AI + Maps" capabilities for customers in retail, mobility, culture and tourism, and other industries.

Zhong Xiangping, Vice President of Tencent Group and President of Tencent Smart Mobility, said bluntly on site: "In the Agent era, the way maps provide services is changing, and the way people connect with the world is being rewritten."

And Tencent Maps is turning maps, location, mobility, and local services entirely into capabilities that can be invoked by AI Agents, helping every kind of intelligence land.

At present, Yuanbao, WeChat Xiaowei, OPPO Xiaobu Assistant, Rokid AI glasses, Guangfan AI full-sensory wearable devices, and others have already integrated Tencent Maps' Agent capabilities.

Tencent Maps releases the full Agent capability landscape! Trip planning can be completed in one sentence

In the Agent era, users no longer translate their needs into simple keywords, but directly state their thoughts: "Help me arrange a weekend trip nearby," "I want to go cycling recently, plan me a heart-shaped route in advance," "a quiet cafe nearby where I can work."

The endpoint of such needs is not a piece of text, but a real place that exists, can be visited now, and is reachable.

To this end, Tencent Maps has released and fully opened up its full Agent capability landscape, turning maps, location, mobility, and local services entirely into capabilities that AI Agents can invoke. At the same time, a series of agents have also emerged, including the Map Card Agent, Search Agent, Guide Agent, and Mobility Service Agent.

Zhang Zhidong, Vice President of Tencent Maps, said that the goal is not just to make maps invokable by AI, but to enable it to understand users, organize solutions, mobilize services, and proactively appear in appropriate scenarios, becoming the location and mobility service foundation for Agents to connect with the real world.

At present, Yuanbao, WeChat Xiaowei, OPPO Xiaobu Assistant, and others have already connected to Tencent Maps' Agent capabilities.

In Yuanbao, route planning can be completed by invoking the map with one sentence, and one-sentence ride-hailing service will also be launched soon.

In September this year, Tencent Maps also co-created a "1+N" agent collaboration model with OPPO ColorOS, jointly releasing the industry's first system-level, memory-capable all-scenario travel agent.

After launch, users went from 9 cumbersome steps to 1 step of directly generating a personalized travel guide with one sentence. Click-through rates in related scenarios doubled, and next-day user retention increased by nearly 1.5 times.

It also supports conversational itinerary adjustments. Planned routes can be directly saved as "My Guide," and starting navigation and sharing with travel companions can be done in one go.

And such map AI capabilities, with the help of agents such as WorkBuddy, Tencent Maps is also opening up to all developers.

Through Agent-friendly interfaces such as MCP Server, Agent Skills, and connectors, developers can give their applications the ability to understand the real world, plan routes, recommend places, and invoke mobility services without adapting from scratch, allowing every user to define their own map.

At present, some developers have developed "How Many More Minutes Can I Sleep," which helps users calculate backward to the latest time they need to leave home, get up, or set off.

It uses Tencent Maps POI data to cross-locate the vague memory fragments of missing persons across multiple dimensions and convert them into candidate sets of geographic coordinates. Push accuracy is expected to increase by more than 40%, and AI clue screening can reduce the clue verification workload of anti-trafficking police by 90%.

Tencent Maps LBSwork released: based on the same-origin Harness framework as WorkBuddy

In fact, the emergence of Tencent Maps LBSwork carries a very simple wish—to let location intelligence truly enter everyone's work and decision-making.

One year ago, Tencent Maps released Gaia 1.0, namely the Spatiotemporal Intelligence Open Platform, an AI productivity foundation. For developers, it makes map development simpler; for platform partners, it turns data and model development into a marketplace; for industry customers, it can provide out-of-the-box agent applications.

One year later, demands from all parties have changed greatly. Based on this, Tencent Maps announced the latest Gaia 2.0. As the full matrix of products and capabilities for Tencent Maps location services AI, Gaia 2.0 includes two product layers: Tencent Maps LBSwork and the spatiotemporal Harness foundation, and ultimately provides efficient solutions for target customers in various industries.

Among them, spatiotemporal Harness lets Agents not only know "how to do the work," but also know "what counts as doing it right."

Li Bin, General Manager of Tencent Maps Location Services, explained that a large model without a spatiotemporal Harness is like an intern without operating procedures: although their ability is not bad, you don't know what they will do next. Take calculating foot traffic for three stores as an example—it might simply add the three numbers together, but those three groups are actually the same people, so adding them creates duplicates. Another example: if you want to open a store by a river, AI analyzes foot traffic within a 500-meter radius for you, but your store is by the river, and customers on the opposite bank cannot easily reach it.

Tencent Maps' spatiotemporal Harness addresses these pain points by accomplishing three things: locking down definitions before work begins, providing full AI oversight during the work, and delivering conclusions backed by evidence at delivery.

Simply put, for developers or industry customers, the underlying frameworks and models can be swapped freely, but business code and data contracts remain unchanged.

The highlight release this time, LBSwork, is a desktop AI map workbench. Built on the same-source Harness framework as WorkBuddy, it deeply integrates Tencent Maps' spatiotemporal dynamic large model, consolidates Tencent Location Services and location big data capabilities along with vertical industry knowledge, and provides one-stop "AI + map" services for customers in retail, culture and tourism, mobility, and other industries.

For different business scenarios, Tencent Maps LBSwork has also launched multiple experts, creating a map brain trust covering seven major directions including retail site selection, marketing insights, crowd heatmaps, and logistics fulfillment.

Take retail site selection as an example. In the past, site selection relied on legwork and reading reports, but now salespeople can easily complete it by simply saying a sentence.

Yu Xiaodong, General Manager of Beijing Leshi Weiguang Network Technology Co., Ltd., feels this deeply: "In traditional industries today, especially the restaurant industry, a single map-based site selection can directly determine life or death. Last year's data showed that the overall restaurant closure rate was 48.9%—for every two stores opened, one does not survive beyond one year."

He said bluntly that traditional experience is not useless; rather, its sample size simply cannot keep up with the speed at which business districts change.

In the past, "guessing, long chains, and poor coordination" were the three fatal knots of traditional site selection. Now, with a single sentence on Tencent Maps, you can generate your own dedicated site selection report. Site selection has officially entered an era of broad equal access, and it has shifted from after-the-fact explanation to before-the-fact calculation.

Currently, brands cooperating with Leshi Weiguang, such as Baodao, Weila, and Xushengji, are all becoming the first batch of experience customers for Tencent Maps LBSwork.

In marketing scenarios, decision-making has also become more scientific and efficient. According to Zhang Xiaoqing, Director of Planning and Operations at Beijing Beichen Shuzhi Technology Co., Ltd., in the past, exhibition organizers had all kinds of data, but could never answer the boss's question: "Who are the people who came?"

Now, for every exhibition and every shopping mall, questions such as "who came, where did they come from, and how much can they spend" can all be calculated clearly through Tencent Maps LBSwork. Merchant recruitment, planning, and post-event review no longer rely solely on experience.

Tencent Maps connects smart hardware! Giving glasses, earphones, and other terminals "hands and feet"

Currently, intelligent services are extending from mobile phones to cars, smart glasses, earphones, and various other terminals. But although they have eyes and ears, they still need hands and feet.

To address this gap, Tencent Maps has now provided continuous, consistent "hands and feet" capabilities for multiple smart terminals.

On the automotive side, Tencent has built a car map Agent. Relying on the coordinated operation of five major Agents—travel planning, map rendering, ecosystem content, intelligent navigation, and parking guides—it delivers high-level in-vehicle travel services.

Users only need to say one sentence to generate a multi-day cross-city family itinerary, and the system can also recommend nearby dining based on personal taste. At the same time, combining vehicle-side perception and lane-level data, the car map Agent can assess complex traffic flows in real time and provide car owners with precise road condition judgments and travel guidance.

In the smart wearable field, Tencent Maps is currently collaborating with Rokid to launch the industry's first native ride-hailing service on AI glasses.

Through Rokid glasses, users only need to say the destination to complete the entire ride-hailing process—from recognizing the request and initiating a ride, to selecting and placing an order, and paying the fare—and can also receive vehicle information and estimated arrival time.

Yang Zhou, head of ecosystem product research at Rokid, said that Tencent Maps has turned payment, maps, mobility, and content—the capabilities that are hardest to develop in-house—into standard components that can be directly called by glasses. Three capabilities are connected through a single call, and this is what an ecosystem truly looks like.

In terms of multi-terminal collaboration, Tencent Maps has also jointly launched the industry's first HarmonyOS smart two-wheeled electric vehicle screen projection navigation solution with Niu Technologies. This solution is deeply adapted to the HarmonyOS system, bringing two-wheeled riders a seamless navigation experience from mobile phone to vehicle screen.

Zhong Xiangping, Vice President of Tencent Group and President of Tencent Smart Mobility, said that in the Agent era, Tencent Maps' original aspiration will not change: "Build the foundation, build connections, build the ecosystem. Tencent Maps will maintain a long-term open posture and work with partners to spread the map wider and thicker."

"Special statement: The above content (including videos, pictures, or audio if any) is uploaded and published by users of "Dafeng Hao," a self-media platform under Phoenix Net. This platform only provides information storage space services.

Notice: The content above (including the videos, pictures and audios if any) is uploaded and posted by the user of Dafeng Hao, which is a social media platform and merely provides information storage space services."