A year ago China’s dexterous-hand field was a fledgling set of demos. Capital interest was thin, most makers stayed at prototype validation, and industry delivery sat in the hundreds-to-low-thousands range, fit for trade shows but not stable industrial use. In one year the ground shifted.
Hot money flooded in. New firms appeared fast; Turning Point reached a 1-billion-US-dollar unicorn just five months after founding, a record for the segment. By IT Juzi’s tally, 2025 funding for the field reached 16.877 billion yuan, and the first half of 2026 alone passed 25 billion yuan, more than all of last year. Shipments crossed from thousands to tens of thousands: over 20,000 units in 2025, with 2026 expected at 70,200.
Technically, hands raced from low to high degrees of freedom across multiple routes, and their role shifted from a robot part to a core terminal of embodied interaction. As the bubble clears and ability diverges, China’s dexterous-hand industry has formed a clear first tier along four complementary routes.
The new four
InTime Robotics, Lingxin, Turning Point and BrainCo make up the new four, each anchored to its own lane. InTime takes the industrial-reliability route, Lingxin the general high-DOF platform, Turning Point a hardware-plus-algorithm manipulation solution, and BrainCo a human-machine bionic fusion route.
The yardstick changed. Out of the lab, a firm’s worth is no longer paper specs but steady delivery, real-scene value and a resilient product matrix. Five criteria now matter: production volume as delivery power, real application as landing power, route breadth as resilience, application fit as ecosystem power, and a freedom-of-motion ladder as matrix power.
Volume separates labs from industry. Building dozens of showcase units is easy; a stable supply chain, standard line and mature yield at ten-thousand-unit scale is the first gate. Scenes separate demos from business: a hand that grabs on a stage is trivial; running steadily on a line or inside a humanoid tests hardware, control and service together, and only real deployment yields the operation data that closes the iteration loop.
Route breadth hedges risk. The three main paths, linkage, tendon and direct drive, each trade off; running several means deeper actuator skill and customisation for clients. Application fit matters because no single universal hand wins: humanoids vary in size, industry in load, and touch sensing is now part of the brief. A freedom ladder, low to high, avoids a price war at one level and covers layered markets.
First tier, harder questions
By these five rulers, InTime, Lingxin, Turning Point and BrainCo became this year’s new four. But a first tier is not a finish line. Low-DOF hands suit standard industrial grabs; mid-DOF balance cost and flexibility; high-DOF serve fine operation, research and premium scenes. The firms that build the full ladder and the broadest real deployments will own the next phase, while single-route players risk being bypassed as the technology converges.
Editor’s note: This is an adapted translation of the original OFweek report. It has been trimmed and restructured for readability for an international business audience.