Meta builds 105-pound humanoid robot prototype to explore future home uses
Meta builds 105-pound humanoid robot prototype to explore future home uses
Meta has developed a 1.5-meter-tall, 48-kilogram humanoid robot prototype named Max in partnership with Fourier to explore internal research on home robotics applications. Built using customized hardware based on Fourier's platforms and NVIDIA's Jetson Thor computing platform, the prototype features five-fingered hands, chest-mounted speakers, and a display face for communication. While Meta aims to improve interaction capabilities like voice recognition and gesture interpretation, the collaboration remains an internal development effort with no plans to sell the robots.
Source: Robotics — Robot learning news discovery · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
Meta builds 105-pound humanoid robot prototype to explore future home uses
Open source S1
Meta has developed a humanoid robot prototype in partnership with Chinese robotics company Fourier to explore potential applications for home robotics. According to a report, the robot was built in just three months, with development of a second version beginning in September. Standing about 4.11 feet (1.5 meters) tall and weighing 105 pounds (48 kilograms), the humanoid reportedly uses NVIDIA’s Jetson Thor computing
Open source S2
named Max, the robot was reportedly developed in three months using customized hardware based on Fourier’s existing robotic platforms. According to The Information, Meta began developing a second version, Max 2, in September, with planned improvements to its hands, torso, and sensing capabilities. The humanoid reportedly uses NVIDIA’s Jetson Thor computing platform, which provides the processing capabilities needed
Open source S4
to support its robotic functions. The project combines Fourier’s humanoid hardware experience with Meta’s interest in developing robots that can interact naturally with people. According to information available, Max incorporates several modifications to Fourier’s existing designs, including changes to its body proportions, head, display, and camera arrangement. The robot also features five-fingered hands designed to
Open source S5
manipulate small objects, including bottles and toys. Chest-mounted speakers provide audio output, while its face can display expressions to support communication. Fourier’s experience with humanoid robots provided a foundation for the project. Its GR-3 humanoid, for example, is designed for close interaction with people and integrates hearing, vision, and touch capabilities. However, Max is not simply a modified GR
Open source S6
-3. The Information describes it as a customized platform drawing on hardware and engineering experience from three generations of Fourier robots. Targets household robotics Meta’s reported design requirements place considerable emphasis on how the robot perceives and communicates with people. These include recognizing spoken commands from a distance of approximately five meters and interpreting gestures, such as wav
Open source S7
ing, from around three meters away. The company also wants the robot to maintain eye contact during interactions and communicate its operating state. This includes indicating whether it is listening, processing information, speaking, or uncertain about a task. These requirements suggest that Meta is exploring ways to make humanoid robots easier to interact with in everyday environments. However, they represent develo
Open source S8
pment targets rather than independently verified capabilities Max has demonstrated. The second-generation prototype, Max 2, is expected to incorporate several hardware changes. These reportedly include a redesigned torso, improved hands, and a projector. Additional sensors are also planned to help the robot navigate stairs and better understand its surroundings. The development effort extends beyond the initial proto
Open source S9
Source:Robotics — Robot learning news discovery · msn.com