Nvidia CEO's Microduck Moment: 20,000 Preorders Make It the Holiday Season's Hottest Toy
A Viral Keynote Demo
Earlier this week, a video of Nvidia CEO Jensen Huang playing with a flock of waddling Microducks during his keynote at the company's GTC Berlin conference went viral. The robots stomped their little feet, made tiny R2-D2 noises, sat, flopped over for a "nap," and somersaulted on voice command. They were so endearing that, by the next day, Microduck preorders had surpassed 20,000 units, according to Hugging Face's Clem Delangue, who shared the news in a post on X.
Positioning as the Holiday Season's Must-Have
Microducks are expected to be delivered in time for Christmas, which means they may already be on their way to becoming the hottest toy of the holiday season. Delangue believes they could end up being the best-selling robot of the year—and not necessarily just for kids. Promotional videos show adults playing with them, too.
Open Source, Built on Nvidia's Stack
Microducks are open source devices produced by Pollen Robotics, a company Hugging Face acquired in 2025. Hugging Face itself was acquired by Nvidia in September, making Microduck an Nvidia product by extension. It is actually Hugging Face's second robot, following the also-cute Reachy Mini.
The Microducks were trained using Nvidia's technology stack, with simulations used to teach them the basic tricks they ship with—walking, sitting, standing, and more. Pollen says owners can train their robots with new tricks and then share them with the community.
Why It Matters
The viral moment offers a welcome palate cleanser after a rough week for the AI industry, in which AI involvement was blamed for cyberattack after cyberattack. AI may have the potential to seriously damage the world—but it can be fun, too. It's hard to watch the demo without wanting to order one.
When @jensenhuang asks for a somersault, you deliver. 🦆
>
See how AI agents on NVIDIA RTX Spark help bring @huggingface's Microduck from simulation to the real world.
>
— NVIDIA RTX Spark (@NVIDIARTXSpark), October 21, 2026
via TechCrunch AI