Meta and Fourier Intelligence Partner to Build Humanoid Robot Prototype Max in Three Months

The robot plan that Zuckerberg has been secretly advancing has finally surfaced.

On October 31, foreign media reported that Meta collaborated with the Chinese robotics company Fourier Intelligence to develop a humanoid robot prototype in only about three months, with the internal codename Max. Currently, a second-generation product has also entered the R&D stage.

This summer, in Fourier Intelligence's factory in Shanghai, this project, unknown to the outside world, was underway.

Li-Chen Miller, Meta's Vice President of Robotics Product, stayed here for nearly two months, working alongside Fourier's engineers to try to turn Meta's vision for robots into reality.

Max is about 1.5 meters tall and weighs 48 kilograms. It is equipped with NVIDIA's Jetson Thor chip, has five-fingered dexterous hands, and its face is a display screen capable of simulating human expressions. According to Meta's requirements, it also needs to recognize human voices and gestures and maintain eye contact with people.

After the news came out, what puzzled us was how these two companies, which usually seem to have little intersection, came together.

After all, there are quite a few humanoid robot companies in China, and Fourier is not the most closely watched one. Conversely, Fourier already has its own robot brand, so why is it still willing to develop products for Meta?

01

Why did Meta choose Fourier?

In February 2025, Meta Chief Technology Officer Andrew Bosworth outlined the company's robotics plan in an internal memo.

Meta not only wants to develop its own consumer-grade humanoid robots, but also to develop AI, sensor, and software technologies that can be provided to other robot manufacturers, maximizing the platform capabilities of Llama.

To put it plainly, Zuckerberg wants to be the Android of the robotics industry.

Google does not need to manufacture all phones itself, yet Android can enter devices from different brands. Meta also hopes that in the future, robots from different manufacturers can all use its AI and software.

Meta must first prove that its technology can truly control a robot.

In September 2025, Bosworth said in an interview with The Verge that the truly difficult part of the robotics industry lies in software, especially dexterous manipulation. Making a robot walk or run is only the beginning; picking up a cup or tidying a desktop is still a complex challenge for robots.

After that, Meta continued to add technical strength in the robotics field. In May 2026, the company acquired the robotics AI startup Assured Robot Intelligence. By the end of October, The Information also disclosed that Meta hopes to make humanoid robots the physical carrier of the AI agent Muse, helping users complete household chores in the future.

It is against this backdrop that Max was born. It is first and foremost an R&D prototype; Meta needs real hardware to verify AI's perception, decision-making, and manipulation capabilities.

So why choose Fourier?

A key detail is that Max is not a direct purchase of Fourier's off-the-shelf product.

According to The Information, the two sides drew on Fourier's R&D experience from its first three generations of humanoid robots and readjusted designs such as body proportions, head structure, cameras, and dexterous hands. Li-Chen Miller, Meta's Vice President of Robotics Product, also stayed on-site in Shanghai for nearly two months to participate in advancing the project.

This means Meta was involved from the hardware design stage, rather than simply installing its software into an off-the-shelf robot.

Fourier's value lies in the fact that it already has experience in developing multiple generations of complete machines. Starting from rehabilitation robots, then to GR-1 and GR-2, and then to GR-3, which emphasizes companionship and interaction, it has a hardware foundation that Meta can continue to modify.

For Meta, this is faster than building a robot R&D system from scratch, and it also better fits its current need to validate AI technology.

Completing Max in three months reflects precisely the efficiency of this mode of cooperation.

Meta is also recruiting supplier development engineers in Shanghai and product design and technical project management personnel in Shenzhen. The related positions involve prototype development, manufacturing, and mass-production preparation. This shows that it is further building robotics hardware R&D and supply chain capabilities.

Of course, Max cannot yet prove that Meta's software can already adapt to robots from different brands. But it has taken an important step: first validating its technology on a deeply customized robot, then exploring the possibility of extending to more hardware platforms.

Meta wants to be the Android of the robotics industry, but before becoming a platform, it must first prove that its technology can truly control a robot.

And for Fourier, what does helping Meta complete this step mean?

In a 2024 interview with TMTPost, founder Gu Jie summarized the company's product philosophy with three words: Made for AI.

This approach gradually became apparent in Fourier's later products and collaborations.

In May 2025, Gu Jie said in an interview that embodied intelligence is not something one company can accomplish alone. Fourier will cooperate with large-model companies according to scenario needs, while continuing to build a stable and reliable robot body.

By early 2026, he further proposed the "1+3+X" ecosystem strategy: health and elderly companionship will be focused on and operated by Fourier, while guided interaction, industrial empowerment, and scientific research innovation will be jointly developed through ecosystem partners. At the same time, Fourier has already adapted its GR series robots to mainstream large models and opened the ActionNet dataset, hoping to promote collaborative R&D across the industry.

From these moves, it can be seen that Fourier does not intend to take on everything from hardware to algorithms to all application scenarios. It would rather do the body well and let different models and development teams work on top of it.

Meta happens to be a special partner. It has its own large models, AI research team, and product ecosystem, but lacks a mature humanoid robot hardware foundation. Fourier, meanwhile, has experience in developing multiple generations of complete machines and hopes its robots can adapt to more models. On the Max project, the two sides formed a relatively clear division of labor.

But this does not mean Fourier has no interests of its own to consider. Cooperating with Meta allows it to engage with the product requirements and R&D standards of a global tech giant, and also gives it an opportunity to bring its hardware capabilities into broader international cooperation. As for the specific R&D revenue, technology sharing, and intellectual property arrangements, the two sides have not yet disclosed them, and it cannot simply be understood as an ordinary order.

There is also a contradiction worth noting here: Gu Jie has always emphasized doing the robot's body well, while what Meta wants to master is precisely the robot's AI and software. Today the two sides can complement each other, but in the future, if Meta gradually acquires more complete-machine R&D capabilities, will the line between partner and competitor change?

This is also a question left by the Max project: as the robotics industry begins to divide labor, how can companies that provide the body avoid ending up with nothing but a body?

03

Conclusion

The emergence of Max has shown the outside world Meta's ambition to enter the robotics industry, and has also drawn renewed attention to the value of China's robotics supply chain.

From motors and reducers to dexterous hands and complete-machine manufacturing, China has already formed a relatively complete robotics industry chain. For humanoid robots that are still in a stage of rapid iteration, the ability to quickly prototype, modify, and deliver is sometimes more important than leading in any single technology.

Meta brings AI and product definition, Fourier provides complete-machine R&D capabilities, and the supply chain supports rapid implementation.

Building a Max in three months reflects both companies getting what they need, and also mirrors the new division of labor taking shape in the global robotics industry.

This article comes from the WeChat public account "Youjie UnKnown," author: Han Lin, editor: Shan Cha, published by 36Kr with authorization.