小米人形机器人完成汽车工厂8小时全轮班作业

Overall: Xiaomi's humanoid robots have completed a full eight-hour shift at an automotive factory, with factory testing underway for eight months.

1. On November 4, 2026, Xiaomi disclosed that two robots completed a full eight-hour shift at a center-console side-cover panel sorting station, matching the pace of full-day production, autonomously handling abnormal deviations, and coordinating with existing production-line systems.

2. First entering the factory in March 2026, they operated autonomously for three consecutive hours at a self-tapping nut loading station, achieving a 90.2% success rate for bilateral installation and adapting to a 76-second takt; by July, the success rate rose to 98%, and the scope expanded to side-cover panel sorting and bin folding and recycling.

Overall: There is also clear progress in mass production and home scenarios.

1. Xiaomi plans small-scale commercialization by the end of 2026, prioritizing automotive manufacturing and campus services, and to deploy more than 2,000 units in its own factories within five years, covering assembly, inspection, handling, and precision testing.

2. A home-service Agent operated continuously for more than 30 minutes in a youth apartment, covering three rooms and completing more than 50 tasks, including tool preparation, tabletop tidying, trash cleanup, item return, and room inspection.

Overall: Xiaomi is using public videos to showcase humanoid robots working on production lines, reinforcing its technology brand image and disclosing the pace of product R&D.

1. In terms of brand communication, on November 4, 2026, it posted a time-lapse video on social platform X, showing two robots completing an eight-hour shift at a sorting station in an automotive factory and visually demonstrating production-line-level operational capability.

2. In product R&D, Xiaomi began in 2021, unveiled the quadruped bionic robot CyberDog and established a robotics lab; in 2022, its first full-size humanoid bionic robot CyberOne debuted; factory testing progressed from three hours, a 90.2% success rate, and a 76-second takt to 98% in July and expansion to more complex stations.

Overall: Its scenario layout spans factories and homes, building product awareness for future commercialization.

1. The mass-production path plans small-scale commercialization by the end of 2026, prioritizing automotive manufacturing and campus services, with more than 2,000 units deployed in its own factories within five years, covering assembly, inspection, handling, and precision testing.

2. For home services, it built an Agent system based on the Xiaomi-Robotics-1 pre-trained model; in a youth apartment, it operated continuously for more than 30 minutes, covered three rooms, and completed more than 50 tasks, indicating R&D layout in home-service scenarios.

Overall: Xiaomi's humanoid robots are moving from factory testing toward small-scale commercialization, signaling demand for scenario deployment and supporting solutions.

1. In terms of market opportunity, Xiaomi plans small-scale commercialization by the end of 2026, prioritizing automotive manufacturing and campus services; over the next five years, it will deploy more than 2,000 units in its own factories, covering assembly, inspection, handling, and precision testing, implying application space across multiple stations.

2. In terms of demand changes, factory testing has expanded from self-tapping nut loading to center-console side-cover panel sorting and bin folding and recycling, and has validated long-duration operation on flexible workpieces; the home Agent covered three rooms and more than 50 tasks, showing an extension from single-point stations to continuous, long-horizon tasks.

Overall: The key lesson is to use production-line takt and exception handling as validation criteria for deployment.

1. In March 2026, three consecutive hours of operation, a 90.2% bilateral installation success rate, and adaptation to a 76-second takt; by July, the success rate rose to 98%, indicating that takt adaptation and success-rate improvement should be addressed first.

2. During the eight-hour shift, the robots could autonomously handle abnormal deviations and coordinate with existing production-line systems, suggesting that commercialization should prioritize stability, exception response, and system integration.

Overall: Xiaomi has disclosed production-line-level validation of humanoid robots in an automotive factory, providing an observable sample for factory automation upgrades.

1. In terms of production needs, the test was conducted at a center-console side-cover panel sorting station, where two robots completed an eight-hour shift, matched the pace of full-day production-line demand, autonomously handled abnormal deviations, and coordinated with existing production-line systems.

2. In terms of station expansion, in March 2026 the robots first operated autonomously for three consecutive hours at a self-tapping nut loading station, with a 90.2% bilateral installation success rate and adaptation to a 76-second production-line takt; by July, the success rate rose to 98%, and the scope covered side-cover panel sorting and bin folding and recycling, completing the first long-duration validation on flexible workpieces.

Overall: Deployment plans point to multiple types of factory stations and large-scale application.

1. Xiaomi plans small-scale commercialization by the end of 2026, prioritizing automotive manufacturing and campus services; over the next five years, it will deploy more than 2,000 units in its own factories, covering assembly, inspection, handling, and precision testing.

2. The implication for factory digitalization is to treat production-line takt, success rate, exception handling, and system coordination as core validation metrics for robots entering factories, rather than judging only by a single demonstration.

Overall: Humanoid robots are moving from single-point demonstrations toward long-duration operation and production-line coordination; service providers can focus on stability and scenario-adaptation solutions.

1. In terms of industry trends, Xiaomi's factory testing has been underway for eight months, progressing from three hours of autonomous operation to a full eight-hour shift, and from a single station to side-cover panel sorting and bin folding and recycling, with the first long-duration validation on flexible workpieces.

2. In terms of technology progress, the home service is based on an Agent system built on the Xiaomi-Robotics-1 pre-trained model; in a youth apartment, it operated continuously for more than 30 minutes, covered three rooms, and completed more than 50 tasks, including tool preparation, tabletop tidying, trash cleanup, item return, and room inspection.

Overall: Customer pain points center on takt, success rate, and exception handling.

1. Factory scenarios require operational pace to match full-day production-line demand; robots need to autonomously handle abnormal deviations and coordinate with existing production-line systems.

2. The data show that raising the success rate from 90.2% to 98% and adapting to a 76-second takt are important validation metrics; service providers can design solutions around takt adaptation, success-rate improvement, exception response, and system integration.

Overall: Xiaomi has publicly disclosed test progress via social platforms and outlined deployment paths for both factory and home scenarios.

1. In terms of release approach, on November 4, 2026, it posted a time-lapse video on X, showing two robots completing an eight-hour shift at a sorting station in an automotive factory, demonstrating production-line-level operational capability and creating shareable technology content.

2. In terms of operational validation, factory testing has progressed for eight months: in March, the robots first entered the factory and achieved three hours of autonomous operation, a 90.2% success rate, and adaptation to a 76-second takt; in July, the success rate rose to 98% and expanded to two complex stations; the home Agent operated continuously for more than 30 minutes in a youth apartment, covered three rooms, and completed more than 50 tasks.

Overall: The commercialization timeline and scenario list can serve as a basis for platforms to observe new hardware categories.

1. Xiaomi plans small-scale commercialization by the end of 2026, prioritizing automotive manufacturing and campus services, and to deploy more than 2,000 units in its own factories over the next five years, covering assembly, inspection, handling, and precision testing.

2. For platform businesses, the article provides information on release cadence, scenario validation, and large-scale deployment plans that can be used to assess the commercialization progress and ecosystem opportunities of the humanoid robot category.

Overall: The Xiaomi humanoid robot case shows an evolution from laboratory demonstrations to production-line-level long-duration operation.

1. In terms of new industry developments, Xiaomi began in 2021 and established a robotics lab; CyberOne debuted in 2022; in March 2026, the robots first entered the factory; in July, the success rate rose to 98% and expanded to complex stations; in November, they completed a full eight-hour shift, with factory testing having been underway for eight months.

2. In terms of new questions, production-line-level validation is not only about single-run success rates but also about adaptation to a 76-second takt, autonomous handling of abnormal deviations, coordination with existing production-line systems, and long-duration operation on flexible workpieces.

Overall: There is an initial framework for the business model and scaling path.

1. Xiaomi plans small-scale commercialization by the end of 2026, prioritizing automotive manufacturing and campus services; over the next five years, it will deploy more than 2,000 units in its own factories, covering assembly, inspection, handling, and precision testing.

2. For home services, it built an Agent system using the Xiaomi-Robotics-1 pre-trained model; in a youth apartment, it operated continuously for more than 30 minutes, covered three rooms, and completed more than 50 tasks, indicating a research direction toward general task capability and multi-scenario reuse.