Hugging Face Unveils Microduck: A $399 Open-Source 25 cm Biped You Train with Reinforcement Learning

Most robotics launches ask you to trust a demo video. Pollen Robotics, the Bordeaux robotics team at Hugging Face, is instead shipping the training loop. This week it opened pre-orders for Microduck, a 25 cm bipedal robot in which every movement — walking, sitting, kicking, roller-skating, standing back up after a fall — is a neural policy trained in a physics simulator and exported to the hardware. It costs $399. The training environments, the reward functions, the domain-randomization settings, and the sim-to-real recipe are all public on GitHub. Microduck follows Reachy Mini, which has shipped more than 10,000 units, but reverses its premise: where Reachy Mini was built to sit on a desk and interact, Microduck is built to leave the desk, fall over, and get back up.

The Hardware

Microduck is 25 cm tall, 14 cm wide, and under 800 g. It carries 15 motors across legs, neck, and head, plus an articulated beak that picks objects off the floor. Compute is a Rockchip RK3566 with an AI accelerator, 1 GB of RAM, and 32 GB of storage.

The sensor stack is unusually complete for the price. A front camera sits behind a dedicated camera-use indicator. Two IMUs are fitted, one in the body and one in the head. Range sensing is a compact LiDAR, an 8×8 time-of-flight matrix. There are microphones and a speaker, two NFC antennas, plus Wi-Fi and Bluetooth. Power is a removable NP-F550 battery, 2600 mAh, good for about an hour.

Seven trained moves ship in the box, driven by a bundled game controller before you write code: walk, sit and stand, kick, grab, roller-skate, and self-recovery. The robot does not speak. Each unit generates its own audio identity on first wake and keeps that voice permanently.