This year, the most active sub-sector in the embodied intelligence field is undoubtedly embodied data—IT Juzi recently compiled a review, and you can check it out via the link provided; the second most popular, in my view, is dexterous hands.
Dexterous hand startups are undergoing intense fundraising—by early August, the dexterous hand sector had recorded 74 funding events involving 47 companies, with a total disclosed amount of approximately RMB 28.51 billion. The total funding for the first seven months has already surpassed the full-year total for 2025, and sector momentum continues to accelerate.
Changes at the industry level are also intriguing—an increasing number of humanoid robot manufacturers are choosing to develop dexterous hands as in-house core modules rather than sourcing them externally. This raises an interesting question: Is the competition in dexterous hands a standalone supply chain segment, or is it becoming an internalized capability of the integrated robot manufacturers?
This report divides the 47 companies into two camps: the “Ontology Group” (15 manufacturers of humanoid robots with in-house developed dexterous hands) and the “Supplier Group” (32 third-party suppliers independently selling dexterous hands and core components), examining their funding data, technical pathways, and competitive dynamics separately.
Our key finding is:
In-house development of the core system has become an entry requirement for original equipment manufacturers, but key underlying components still rely heavily on external suppliers;
The supplier segment is experiencing a structural divergence, with distinctly different moats and competitive landscapes across the three subcategories.
Chapter 1: The 2026 Financing Landscape for dexterous hands: "Ontology Group" and "Supplier Group" Operate as Independent Systems
From January 2026 to early August (data as of August 3), the IT Juzi database recorded 74 funding events in the dexterous hand sector, involving 47 independent companies.
When extended over a longer time horizon, the funding interest in the dexterous hand sector shows a steep upward curve:
In terms of the number of incidents, 2024 more than doubled compared to the previous year, and 2025 nearly tripled compared to 2024.
In terms of amount, the growth is even more striking: 28.51 billion yuan in the first eight months of 2026 has already surpassed the full-year total of 16.9 billion yuan in 2025.
With nearly five months of data still unaccounted for in 2026, the annual total is likely to exceed RMB 40 billion. The dexterous hand sector is rapidly gaining momentum.
Notably, while the number of incidents surged in 2025, the average per incident declined (16.9 billion ÷ 99 incidents ≈ 170 million per incident), indicating a flood of early-stage projects that year. In contrast, during the first eight months of 2026, although the number of incidents was slightly lower than the same period in 2025, the average per incident rose to 385 million yuan, signaling a shift from "quantity expansion" to "concentration of funds"—capital is increasingly flowing toward top-tier projects.
It’s not appropriate to compare companies that build both ontologies and dexterous hands with those that focus primarily on dexterous hands when evaluating funding.
We categorize them into two groups based on business attributes, looking at those that raised funding in 2026:
The companies mentioned below are only those with funding records within 2026 and do not represent all companies in this sector.
Original equipment manufacturers (15 companies): Core business is the production of humanoid robots, with dexterous hands developed as proprietary core modules—not sold separately externally, or even if capable of external supply, currently used primarily for their own products.
These 15 companies include ZhiJi Dynamics, StarMotion Epoch, Galaxy General, Independent Variable Robotics, Stardust Intelligence, Zhongqing Robotics, Magic Atom, Matrix Superintelligence, Yuequan Bionics, Qingbao Robotics, ZeroPower Robotics, Zhejiang Humanoid, Songyan Power, Gesong Technology, and HalfAwake Embodied.
In the first seven months of this year, they collectively completed 23 funding rounds, with disclosed amounts totaling approximately RMB 18 to 20 billion.
Third-party suppliers (32): Independently develop and sell core components such as dexterous hands, haptic sensors, micro motors, and precision ball screws, serving customers including humanoid robotics companies, research institutions, and industrial automation manufacturers.
These 32 companies completed a total of 51 funding rounds in the first seven months of this year, with disclosed amounts totaling approximately RMB 8.51 billion.
When comparing the two sets of data, three observations stand out:
First, the original equipment manufacturer's "per-transaction amount" far exceeds that of the supplier.
The original equipment manufacturer has raised 23 rounds of funding, totaling approximately RMB 20 billion, with an average of over RMB 800 million per round; suppliers have raised 51 rounds of funding, totaling RMB 8.5 billion, with an average of less than RMB 200 million per round.
This indicates that the original equipment manufacturer is valued using a "whole-vehicle" valuation logic, where investors are betting on the entire humanoid robotics sector, with dexterous hands serving merely as one technological pillar. In contrast, suppliers are valued using a "component-level" valuation logic, which naturally has a lower ceiling.
Second, the number of suppliers of dexterous hands is nearly twice that of body manufacturers.
This means the entry barrier for startups focused on dexterous hands as a standalone category is lowering—core teams of 20 to 30 people, along with several million yuan in funding, are sufficient to get started. However, it also means competition in the supplier segment is far more intense than in the end-product segment.
Third, the funding pace of the two camps is completely different.
The financing of protocol companies is concentrated in Series B and later stages, with over 56% of events occurring after Series B, some even reaching Series C and Pre-IPO. In contrast, suppliers show high concentration in the Angel to Series A stages, with over 45% falling in the Angel to Pre-A range. In other words, protocol companies have entered the "mid-to-late" stage of financing, while suppliers remain in the "early-to-mid" stage.
Chapter 2: Fifteen OEMs Have Made Self-Developed Dextrous Hands a "Standard Practice"
The landscape for dexterous hands has changed significantly among original equipment manufacturers; in-house developed dexterous hands have shifted from being a "differentiation strategy" adopted by a few leading companies to a "barrier to entry" for full humanoid robot manufacturers.
The table below lists basic information for the 15 robot manufacturers that raised funding in 2026. We have selected three—ZhiJi Dynamics, Xingdong Jiyuan, and Xingchen Intelligence—to represent three distinct technical pathways: industrial applications, all-electric direct drive, and cable-driven systems, for a comparative in-depth analysis.
Overall, the 15 robot manufacturers that raised funding this year have clearly chosen to develop dexterous hands in-house—the rationale is straightforward: the dexterous hand is the core organ enabling robots to perform tasks, and outsourcing it means surrendering the most critical differentiating capability to others. In this early stage of the industry, where technological pathways have yet to converge, an increasing number of emerging robot manufacturers are unwilling to take this risk.
However, one notable signal is that mainstream manufacturers primarily focus their in-house development on the humanoid hand itself (mechanical structure and drive control), while most still source foundational components—such as tactile sensors, micro motors, and precision ball screws—from external suppliers. The technological barriers and capital intensity of these specialized segments remain difficult for integrators to overcome in the short term. This is why suppliers in these niches continue to attract concentrated investment this year.
Zhuji Power: High-Load Solutions for Industrial Scenarios
Zhuji Power has chosen a "scenario-defined product" approach. Its proprietary three-finger dexterous hand is designed for use with the full-sized humanoid robot CL-2, focusing on high-load grasping in industrial scenarios. The three-finger design strikes a balance favoring load capacity over flexibility, making it ideal for heavy-duty tasks such as material handling and assembly on factory production lines.
This year, the company completed a Series B round of $200 million and a Pre-IPO round of $200 million, totaling over RMB 2.8 billion, ranking among the top in funding volume among 15 blockchain infrastructure providers. The occurrence of the Pre-IPO round indicates that the company is preparing for an IPO, signaling a clear shift into the later stages of its fundraising journey.
Star Motion Era: Exploring the Technical Limits of Full Direct Drive
Star Epoch takes a fully direct-drive approach, with its proprietary XHAND series of dexterous hands. The XHAND1 features 12 active degrees of freedom, can lift objects weighing up to 25 kg, and performs over 100 complex tasks. The benefits of the direct-drive solution include fast response and high control precision, but it also presents the greatest challenges in cost and size among the three options—choosing this path means tackling difficult engineering hurdles.
This year, the company completed strategic investments of over RMB 1 billion and USD 200 million, along with a Series B round of RMB 1 billion, placing its fundraising scale firmly in the top tier. From the pace of financing, StarMotion is rapidly turning its technological leadership into commercial advantages.
Stardust Intelligence: The Production Pioneer of Cable-Driven Solutions
Star Dust Intelligence adopts the world's first "cable-driven" technology approach, independently developing all components from full-body actuation to dexterous hands. The cable-driven design replaces rigid transmission with tendons, offering a biomechanical structure closer to the human hand, resulting in greater flexibility and human-like motion—though ensuring long-term reliability and precise control remains an engineering challenge.
Notably, Xingchen Intelligence has achieved mass production and delivery of over a thousand units of the S1, with the T1 starting at RMB 89,900—making it the first among the three technical pathways to bring pricing into a range perceptible to consumers. The company completed its B-round financing of RMB 1 billion this year.
Chapter 3: Typology of Smart Hand Suppliers: The Three Types in the Industry
The 32 companies currently involved in the supply of dexterous hands (with funding as of 2026) can be broadly categorized into three groups. This classification is not academic but rather based on business logic—each group serves different customers, possesses distinct competitive advantages, and faces entirely different competitive landscapes.
Category 1: Full-hand manufacturers ("those who sell hands")
The core product of these companies is a complete dexterous hand—integrated from fingers and joints to drive and control—packaged as a plug-and-play module sold to companies building humanoid robots, research institutions, or industrial automation integrators. Their competitive advantage lies in systems engineering capabilities: hardware design, mass production processes, cost control, and seamless integration with clients’ robotic control systems.
This year, several unicorns have emerged in this category. Lingji split off from Zhiyuan less than six months ago and is now valued at ten billion yuan; Lingxin Qiaoshou, founded in 2023, has reached a valuation of 15 billion yuan and is already preparing for a Hong Kong IPO; Xinuo Future, established one and a half years ago, has surpassed a $1 billion valuation.
A consensus is forming in this space: dexterous hands will become one of the most critical and highest-value standalone modules in humanoid robots, analogous to engines in internal combustion vehicles or batteries in electric vehicles.
But competition has become visibly intense—according to IT Juzi, nearly 20 companies claim to be developing whole dexterous hand systems, ranging from independent manufacturers like Yinsi Robotics and Lingqiao Intelligence, to emerging players such as Yuan Sheng Intelligence, Zhixing Robotics, Wan Na Robotics, and Boya Intelligence, as well as specialists like Zhongke Guiji, Xingji Guangnian, and Huiling Technology, each vying to advance their technical approaches, accelerate mass production, and secure customer partnerships.
Critical point AGILINK
Full name: Shanghai Critical Point Innovation Intelligence Technology Co., Ltd. | Established: January 2026 | Headquarters: Shanghai
Team: Founder Xiong Kun, who holds a Master’s degree in Electronic and Information Engineering from the Robotics Institute at the Hong Kong University of Science and Technology. He joined Tencent’s Robotics X Lab in 2018 during its early stages, contributing to its foundational development and leading robot hardware and software architecture. He later worked at the Institute for Advanced Study (IDEA) and Inovance Technology, and joined AgiBot in November 2024 to lead its dexterous hand business. Co-founder Deng Taihua is Chairman of AgiBot, and Wang Chuang previously served as President of AgiBot’s General Business. The team brings together top-tier research, industry giants, and startup expertise.
Funding Journey: Seed round in January 2026, Series A in February, Series A+ of hundreds of millions in May, and Series B of 1 billion in June, with valuation surpassing 10 billion. From registration to unicorn status, achieved in just five months.
Key highlights: A flagship project incubated by the Zhiyuan system, starting at an exceptionally high level—profitable from inception. In less than six months, over 8,000 OmniHand dexterous hands and more than 10,000 grippers have been delivered. Its commercialization capability is currently the strongest among pure dexterous hand companies. Xiong Kun’s strategic insight: The competition for dexterous hands is not in the lab, but on the production line.
Skillful and ingenious hands
Full name: Lingxin Qiaoshou (Beijing) Technology Co., Ltd. | Established: July 2023 | Headquarters: Beijing
Team: Founder, CEO, and CTO Zhou Yong, a graduate of the Young Talent Program at Huazhong University of Science and Technology, brings over 15 years of combined experience in internet and robotics, with two prior full-cycle entrepreneurial ventures (3D simulation and intelligent autonomous vehicles), and has led major national projects such as the Hong Kong-Zhuhai-Macau Bridge. Co-founder Zuo Jiaping comes from Daata Robotics and Ninebot Robotics, overseeing hardware mass production and supply chain management. Chief AI Architect Su Yang leads the development of cloud-based large models and data infrastructure. The R&D team comprises over 100 members, with more than 90% of core components自主研发 and produced in-house.
Funding history: Seed round in April 2025, angel round in August 2025, Series A in October 2025, Series A+ in November 2025, Series A++ in December 2025; Series B of RMB 1.5 billion in February 2026, Series B+ in April 2026, with a valuation of RMB 15 billion. Rumors indicate that preparations for a Hong Kong IPO have already begun.
Key highlights: The company holds over 80% market share in the global high-flexibility dexterous hand segment, with monthly production exceeding 4,000 units and comprehensive coverage of three core technical pathways: linkages, direct drive, and tendon-driven systems. Zhou Yong is an intriguing individual—graduated from the gifted program at Huazhong University of Science and Technology, he previously built a gaming community (gaining foundational expertise in 3D simulation), developed autonomous vehicles (mastering hardware engineering), and ultimately integrated all his prior experience into dexterous hand technology. He says his ultimate motivation for starting this venture stems from a childhood obsession with the anime "Doraemon."
Xynova Future
Full name: Hangzhou Xinuo Future Technology Co., Ltd. | Established: December 2024 | Headquarters: Hangzhou
Team: Founder and CEO Xia Yuxuan, a graduate of Columbia University, previously worked at leading international private equity firms and investment banks including鼎晖投资, Morgan Stanley, and Warburg, with deep investment exposure across the robotics industry value chain. This background—transitioning from investor to entrepreneur—is rare in this space, giving him a natural advantage in judging capital cycles and industry inflection points. The team brings together top talent from elite institutions such as Tsinghua University, Zhejiang University, Shanghai Jiao Tong University, Columbia University, and the University of Pennsylvania, alongside industry experts from DJI, Schaeffler, KUKA, Apple, and CATL.
Funding History: Seed round in December 2025, Pre-A round in March 2026, A round in May 2026 for hundreds of millions of yuan (led by Xiaomi, Li Auto, CITIC Construction Capital, etc.), A+ round in July 2026, with valuation surpassing $1 billion, officially becoming a unicorn.
Key highlights: Creator of Flex 2, the world’s first hybrid manipulator combining cable-driven and direct-drive technologies. Xia Yuxuan once made a lasting impression with a remark: during a visit to a charger factory, he saw a female worker spending her entire day doing nothing but removing labels from products. This image became the catalyst for his entrepreneurial journey—what if a sufficiently dexterous hand could replace such repetitive tasks? Xiaomi has invested continuously from the angel round through to the A+ round, and the strong backing from industrial capital confirms that the product has passed validation by automotive manufacturers. The company is projected to achieve an annual production capacity of ten thousand units by the end of 2026.
Category Two: Haptic Perception Solution Providers ("Those Who Make the Hands Feel")
For a robot to truly work effectively, having dexterous fingers alone isn’t enough. Without tactile feedback, it won’t know how much force to apply when picking up an egg or whether a glass bottle might slip away—dexterous hands without touch are merely expensive props.
In the field of haptic sensors, overseas manufacturers have long dominated the market (with individual units priced at over 100,000 RMB), but this year, several domestic companies are rapidly changing the landscape.
A common characteristic of these companies is extremely high technical barriers and strong integration with downstream partners. Once a robot manufacturer has adapted a specific haptic solution, the cost of switching is very high—meaning that first-mover advantage in this space can be accumulated over time.
Pacini
Full name: Pacini Artificial Intelligence Technology (Beijing) Co., Ltd. | Established: June 2021 | Headquarters: Beijing
Team: Founder and CEO Xu Jincheng, who holds a Ph.D. from Waseda University in Japan and studied under Professor Shigeki Sugano, a leading authority in Japanese robotics. His advisor, Ichiro Kato, was the inventor of WABOT-1, the world’s first humanoid robot in 1973. In terms of academic lineage, Xu Jincheng can be regarded as the third-generation heir to the “authentic school” of humanoid robotics. Chief Strategy Officer Luo Xiaoheng is responsible for strategy and fundraising.
Funding History: Angel round in November 2021 (Ji Qi Chuang Tan), Pre-A round in December 2022, Series A and A+ rounds in 2024, two additional A+ rounds in 2025, followed by a B round of RMB 1 billion in March 2026 and a strategic investment of RMB 1 billion in August, reaching a valuation of RMB 15 billion. Over ten years, the company has completed ten funding rounds with steady pacing.
Key highlights: The undisputed market leader in China’s haptic sensor sector, with proprietary ITPU multidimensional haptic sensing technology that leads globally—some modules have already entered the three-digit price range, reducing costs by two orders of magnitude compared to overseas counterparts priced at around 100,000 RMB. The company follows a “from inside out” trajectory, progressing from sensors to dexterous hands to full humanoid robots. BYD and JD.com have both transitioned from customers to shareholders, a strong signal in the manufacturing sector. Xu Jincheng’s dedication to haptics borders on belief—he says, “It’s not the brain that commands the hand; it’s the hand that shapes the brain.”
Dai Meng Robot
Full name: Daimeng (Shenzhen) Robotics Technology Co., Ltd. | Established: December 2021 | Headquarters: Shenzhen
Team: Founded by Professor Yu Wang, founding director of the Robotics Institute at the Hong Kong University of Science and Technology, and his student Dr. Jianghua Duan. Professor Wang has spent decades specializing in robotic manipulation and is widely recognized as an authoritative scholar in the field. The team has deep collaborative ties with Professor Edward Adelson of MIT (inventor of the GelSight vision-tactile technology), having fundamentally improved MIT’s technical approach.
Funding History: Angel round in September 2023, Angel round in November 2024, Angel+ round in August 2025, strategic investment of RMB 100 million in December 2025, Series A round of nearly RMB 100 million in June 2026 (led by Huichuan Industrial Investment and China Telecom), valuation of RMB 3 billion.
Key highlights: The first company in China to achieve autonomous mass production of visuotactile sensors, featuring 40,000 sensing units per square centimeter—far exceeding the density of human fingertips. In April 2026, the company open-sourced over 10,000 hours of tactile data, surpassing one million downloads in the first month and topping the charts as the leading embodied dataset for the physical world. What sets this company apart is that it is not selling tactile sensor hardware—it is building “tactile infrastructure.” The decision to open-source the data reveals Professor Wang Yu’s focus on cultivating an industry ecosystem, not just short-term gains.
Thousand Awareness Robotics
Full name: Qianjue Robotics Technology (Shanghai) Co., Ltd. | Established: May 2024 | Headquarters: Shanghai
Team: Founder Ma Daolin holds a bachelor’s and doctoral degree in Mechanics from Peking University and completed postdoctoral research at MIT’s Mcube Lab. He is currently an associate professor and doctoral supervisor at the School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiao Tong University. In 2021, he received the sole Best Paper Award at ICRA, the premier conference in robotics, for proposing a novel approach to robotic manipulation based on tactile sensing for spatial perception. Co-founder Liu Jin holds a Ph.D. in Mechanical Engineering from Shanghai Jiao Tong University and oversees company operations; Zhao Haonan leads the development of sensor hardware底层. Core team members come from top institutions including MIT, Peking University, Tsinghua University, and Shanghai Jiao Tong University.
Funding Journey: Seed round in October 2024, Seed+ round in May 2025, Pre-A round in October 2025, and Series A round in July 2026. Four rounds in two years, serving over 300 clients.
Key highlights: The world’s first micro multi-modal haptic sensing system establishes a complete technological closed loop encompassing “haptic sensors, data acquisition, simulation, VTLA model, and scenario solutions.” Madao Lin exemplifies the path from top-tier academic papers to industrial application—a trajectory increasingly recognized in the hard tech sector. Qianjue’s VTLA model integrates haptics and vision at the model level, rather than merely combining sensors.
Category Three: Core Drive and Transmission Component Manufacturers ("The People Who Make the Hands")
If the first two types of companies are the stars on stage, this type of company is the "hidden champion" behind the scenes. They don’t sell complete dexterous hands, but the most critical components of a dexterous hand—motors, gearboxes, and ball screws—may originate from them.
These companies are the most stable in the long term, because regardless of which technology path ultimately prevails for dexterous hands, motors, ball screws, and reducers remain essential components. They benefit from the advantages of “domestic substitution” and “technological downgrading”: on one hand, breaking foreign monopolies; on the other, using process innovations to reduce costs to one-tenth or even less of the original level.
Wuji Technology
Full name: Shanghai Wujī Technology Co., Ltd. | Established: March 2019 | Headquarters: Shanghai
Team: Founder Pan Yunzhe graduated in 2018 from the University of Illinois Urbana-Champaign (UIUC) with dual degrees in Computer Science and Chemistry. His motivation for starting the company was simple—he was dissatisfied with the heavy and underpowered motors available on the market, and spent three years developing the F-series motor from scratch.
Funding history: Angel round in August 2023, Pre-A round in November 2023, A round in November 2025, A+ round in February 2026, and multiple consecutive A+ rounds completed intensively from March to July 2026, with a stellar lineup of shareholders.
Key highlights: The F-series motor delivers a peak torque of 8 Nm at just 200 grams, achieving a power density that dominates its class and places it among the top tier globally. The recent surge in A+ funding rounds reflects strong investor confidence, while downstream orders are skyrocketing, necessitating continued capacity expansion.
Fingertip Intelligence
Full name: Suzhou Zhi Jian Zhi Qing Technology Co., Ltd. | Established: August 2024 | Headquarters: Suzhou
Team: Founder and CEO Zhang Yang, who earned his undergraduate degree in Electrical Engineering and Automation (Motor Direction) from Central South University, brings 12 years of experience in the motor industry, having previously held key roles at Midea Group, Weifu High-Technology, and Innovusion, where he led the development of the world’s first 1550nm automotive LiDAR scanning system with annual sales exceeding 150,000 units, demonstrating full-stack capabilities from product definition to mass production. CTO Han Jie has 10 years of motor R&D experience and previously led the mass production of multiple main drive motors and precision motors at Geely and Innovusion. The founding team is further strengthened by top scholars from CMU providing technological foresight.
Funding history: Angel round in August 2025, Pre-A round in June 2026—both rounds perfectly align with the timeline from technical validation to commercial validation.
Key highlights: The world’s first mass-production process for PCB axial flux stator windings—applying semiconductor manufacturing principles to motor design, resulting in a 48-layer PCB stator with a mere 4.2 mm thickness, completely revolutionizing traditional hand-wound micro motor manufacturing. With a minimum outer diameter of 4 mm, it is the smallest commercially mass-produced hollow-cup motor globally. Mass production yield has reached nearly 100%, with costs over 30% lower than traditional solutions. If this technology pathway continues to evolve, it could redefine the rules of the micro motor industry.
Noshi Robot
Full name: Shenzhen Nuo Shi Robotics Co., Ltd. | Established: July 2023 | Headquarters: Shenzhen
Team: Founder Xu Yang holds a master’s degree in mechanical manufacturing from Tongji University and previously served as head of the R&D department at Valeo’s Asia-Pacific R&D center. Co-founder Wang Xiaobin was formerly vice general manager of a Valeo subsidiary and is a serial entrepreneur who built a team from scratch to the Pre-IPO stage. Chief technical advisor Professor Xu Genlin, a retired professor from Shanghai University, has spent nearly 50 years specializing in mechanical manufacturing processes—he is Xu Yang’s father and the technological foundation of the company. This company is truly a story of two generations passing the torch: the older generation accumulated half a century of expertise in lead screw manufacturing, while the younger generation has industrialized it through modern corporate organization.
Funding history: Seed round in December 2024, Seed+ round in April 2025, and a Series A round of hundreds of millions in March 2026.
Key highlights: The world’s smallest planetary roller screw—1.5 mm in diameter and 5.5 mm in nut size, thinner than a toothpick, yet capable of delivering a stable load force of 10 kg with a lifespan exceeding 3 million cycles, priced under 100 yuan (industry sample prices typically exceed 10,000 yuan). Through proprietary high-precision metal forming technology, production efficiency has been increased tenfold. Currently, humanoid robotics and automotive steer-by-wire chassis applications each account for half of the business, providing a balanced dual-engine structure with greater flexibility to adapt to industry rhythms.
Chapter 4: Divergence, Convergence, and Irreversibility in the Dextrous Hand Sector
The dexterous hand sector is undergoing a structural分化.
The hybrid model of "in-house development + external procurement from suppliers" will become the industry standard.
Funding data for 2026 shows that all 15 robot manufacturers have clearly developed their own dexterous hands, collectively securing approximately RMB 20 billion in funding—this could be interpreted as negative for suppliers. However, the reality is not a simple binary opposition.
The key point is that while mainstream manufacturers focus their proprietary development on the humanoid hand itself—such as mechanical structures and drive control—they still rely heavily on external suppliers for core components like tactile sensors (Pacini, Daimeng, Qianjue), micro motors (Wuji Technology, Zhijian Zhiqing), and precision ball screws (Nuoshi Robotics).
The technical barriers and mass production know-how for these components are not automatically yours just because you’re a system integrator; in fact, companies that assemble complete systems often struggle with making components—that’s a hard truth in manufacturing.
The industrial landscape of dexterous hands is unlikely to be fully internalized by OEMs, as seen with power batteries; instead, it is more likely to resemble automotive electronics, with tiered competition between Tier 1 (integrated dexterous hand modules) and Tier 2 (core components), each occupying distinct ecological niches.
The original equipment manufacturer receives a "sector valuation," while the supplier receives a "product valuation"—these two are not comparable.
Of the 74 funding events in 2026, ontology manufacturers secured 23 deals totaling approximately RMB 20 billion, while suppliers secured 51 deals totaling approximately RMB 8.5 billion—the average deal size for the former was more than four times that of the latter. However, this does not imply that suppliers have lower investment value.
The main difference lies in their valuation logic: the former reflects expectations for the entire embodied intelligence sector, betting on the direction; the latter reflects the market share and revenue growth of a specific product, betting on certainty.
Over the next two years, as ontology manufacturers enter the IPO or Pre-IPO stages, their valuations will face a rigorous test of performance; meanwhile, suppliers that achieve profitability and scale may end up more stable than many original equipment manufacturers.
The supplier segment will undergo a brutal consolidation, with integrated machines from Lingqiao Shou being the first to feel the impact.
Nearly 20 out of 32 suppliers claim to be manufacturing dexterous hands as complete units—a figure that speaks volumes. Dexterous hands as complete units are a category with strong economies of scale: higher production volume, higher yield, and lower unit cost. Without sufficient shipment volumes, smaller players cannot cover the amortization of molds and depreciation of production lines, which can easily overwhelm them.
Critical Point and Lingxin Qiaoshou are already overwhelming competitors with their production capacity and pricing. Xino’s future orders are highly certain, backed by Xiaomi. By the end of 2027, the number of viable players in the dexterous hand product category may significantly decline.
The "hidden champions" of haptic perception and core components will enjoy the longest period of benefits.
Unlike the complete dexterous hand system, in the two specialized areas—tactile sensors and micro motors/lead screws—each leading company has built a 3- to 5-year first-mover advantage around a highly vertical technical niche.
Pacini has over five years of accumulated expertise in haptic sensing; Daimeng’s open-source data assets exhibit network effects; Wujī Technology’s motor power density is difficult to replicate in the short term; Nuo Shi’s micro lead screw technology benefits from nearly 50 years of family-held know-how.
The common traits of these companies are: the market niche is small enough that large players aren’t interested in entering, the technology is sophisticated enough that new entrants struggle to catch up quickly, and the customer base is concentrated enough that once locked in, clients are hard to replace. While they may not become billion-dollar platforms, their likelihood of becoming successful, “small but beautiful” companies valued between hundreds of millions and several billion dollars is extremely high.
This article is from the WeChat official account "IT Juzi" (ID: itjuzi521), authored by Wu Meimei.