Gasgoo Munich-On November 10, Vbot officially launched the DEV Geek Edition, a high-performance robot dog and quadruped development platform designed for individual developers and tech enthusiasts. The official price is 22,988 yuan, with a limited-time Double 11 promo dropping it to 20,988 yuan.
This marks the first time Vbot has pulled a flagship quadruped development platform below the 20,000-yuan threshold — and it’s squarely aimed at developers, not end consumers.
Image credit: Vbot
Fully Equipped for Under 20,000 Yuan: Lowering the Barrier to Entry
Vbot’s move into quadruped development platforms is no spur-of-the-moment decision.
The company’s very first product was a quadruped robot dog named "Datou." Launched in December 2025, it secured 6,540 pre-orders — pulling in nearly 100 million yuan.
Mass deliveries began this May. By cumulative sales, it now ranks as the world’s top-selling consumer-grade robot dog.
The new DEV Geek Edition builds directly on that market-proven quadruped platform, tailored specifically for developers.
For individual developers, the experience hasn't been great.
Quadruped robots that support secondary development typically start between 30,000 and 50,000 yuan. Fully loaded models with robust sensors and computing power run 40,000 to 90,000 yuan — roughly the down payment on a family car.
Worse, many ship as bare-bones units. Developers have to add their own LiDAR and build their own environments. By the time everything is configured and dependencies installed, the initial enthusiasm has often worn thin.
Vbot’s approach: bundle all the hardware upfront, with every component mapped to a specific development scenario.
Image credit: Vbot
Specifically, the DEV Geek Edition comes with a 360° LiDAR for mapping and navigation, an HD binocular camera for visual recognition and distance measurement, and a microphone array for voice commands. Add a screen, speaker, and ear lights for interaction, and developers no longer need to buy basic sensors separately — boot up, and the robot can perceive its surroundings and understand commands right away.
Connectivity spans Bluetooth, Wi-Fi, and 4G, freeing developers from ethernet cables and allowing outdoor debugging. A full-featured Type-C port supports external cameras, speakers, and projectors, while physical expansion screw holes make it easy to mount backplates, cargo baskets, and DIY accessories — leaving ample room for custom mods.
Hardware aside, software dictates development efficiency.
The DEV Geek Edition offers six core functional modules covering spatial memory and navigation, follow mode, local visual recognition, voice interaction, the RCP robot control platform, and a motion-pack repository. With over 50 preset actions and expressions, developers can call and combine them directly — no need to code from the ground up. It essentially eliminates the need to reinvent the wheel.
The base layer is compatible with the mainstream ROS 2, allowing seamless migration for developers familiar with that system. Vbot also open-sources its self-developed distributed middleware, Aorta, which handles communication and scheduling across multiple devices, leaving headroom for performance and expansion.
Even more aligned with current development habits are its AI-native tools.
The platform provides an SDK, documentation, a simulation Studio, and an AI Agent. Developers simply describe their needs in natural language, and the AI helps query interfaces, write code, and debug.
Hardware Can Be Copied; Ecosystem Stickiness Is the Real Moat
A platform is only as good as the community it attracts.
At the SheNicest hackathon this August, teams from diverse backgrounds already built working projects on the Vbot robot dog. "Lingling," a hospital companion bot, uses autonomous navigation and voice interaction to guide patients and remind them of appointment steps. "Jiajiagou," a home fitness buddy, proactively invites users to exercise and uses pose recognition to coach yoga sessions.
Image credit: Vbot
To sustain this collaborative ecosystem, Vbot is pushing a "1,000-Person Co-Creation Plan," recruiting 1,000 embodied AI developers worldwide to run their own algorithms on the platform. It’s also integrating ecosystem resources like Digua Robotics to help developers connect with hardware and commercialization channels.
Vbot isn’t alone in this push. From Apple open-sourcing its HALO robot control system to other vendors gradually unlocking SDKs and simulation tools, the industry’s competitive focus is shifting from selling hardware to competing on platforms and ecosystems.
After all, hardware can be replicated by latecomers, but the code, applications, and habits developers accumulate on a platform create switching costs.
Yet the launch price is just the gateway. Beyond the 1,000-person co-creation push, what truly matters — and warrants closer tracking than price — is how many projects and applications will survive and genuinely enter everyday life.
