China’s industrial control giants are having a better year. Inovance reported revenue of RMB 37.956 billion for the first nine months of 2026, up 19.88 per cent, with net profit attributable to shareholders of RMB 4.398 billion, up 3.38 per cent.
The headline profit number is unremarkable. The line underneath it is not. General automation, the servos, inverters and programmable logic controllers that indicate how busy Chinese factories are, brought in about RMB 15.9 billion, up 33 per cent. Emerging businesses grew 88 per cent.
Then there is the company the industry calls the small Inovance. Leadshine, formally Shenzhen Leadshine Technology, reported first-nine-month revenue of RMB 1.909 billion and net profit of RMB 299 million, up 46.77 per cent and 86.8 per cent respectively. It has now posted positive growth for seven consecutive quarters.
A twenty-fold size gap and the opposite growth rate
The two companies differ in scale by a factor of roughly twenty, and the gap is not closing. What separates them is where each one chose to fight.
Inovance built its position on servos and won it. By Ruigongye’s count, it held 35.2 per cent of China’s general servo systems and 20.6 per cent of low-voltage inverters in the first half of 2026, first in both. On the speed, torque and responsiveness of a motor on a production line, it has real pricing power.
The one hill it has not taken is the programmable logic controller, the industrial computer that sequences hundreds of actions on a line and decides when a temperature excursion triggers an alarm. In the first half, Inovance held about 6.7 per cent of China’s mid-range and large PLC market, third place. The leader is a foreign company with 51.5 per cent.
That gap explains why servos were localised first. Servo localisation rode a hardware curve: IGBT localisation passed 50 per cent, upstream costs fell by about 30 per cent, and Chinese suppliers took share through cost and response time. PLCs are not decided by a component. They are decided by software ecosystems, by the programming habits engineers have built over decades, and by accumulated engineering knowledge. Inovance has said large PLCs are a strategic product and is trialling them in mining and port applications.
Going at the joint instead
Leadshine did not walk into that fight. It picked points where a smaller company can win a segment rather than a market.
In servo systems it grew revenue 66.5 per cent in the third quarter and holds second place among domestic brands. In humanoids it concentrated entirely on a few components: frameless torque motors, joint modules and dexterous hands.
The order book is the evidence. Frameless torque motor deliveries exceeded 120,000 units in 2025. By the end of the first half of 2026, orders in hand had passed one million units, with 800,000 to 900,000 expected to ship during the year. Monthly capacity has been pulled up to 150,000 units, and the company is planning for three million a year.
Harmonic and planetary joint modules passed 60,000 units of orders. Dexterous hands have entered small-batch delivery. Robot-related orders for the full year are expected to land between RMB 200 and 300 million. On the A-share robotics component board, that puts the company in the first tier.
The bet underneath the numbers
The logic is not complicated, and it depends on one assumption about how humanoids will be built.
If the industry follows the route Tesla has set out, joints and hands reach volume first, and general-purpose capability arrives later as the models improve, then the first reliable revenue in the chain belongs to whoever can hold actuator yield and cost. Whole-machine companies will keep buying capacity and reporting losses while their suppliers book deliveries.
That is why Leadshine declined to contest the PLC, the brain of the factory, and went instead for the joints and the muscles. For a specialist of its size facing a platform incumbent, attacking the highest-value component on the robot is more realistic than attacking the controller that runs the building. It is also a smaller prize, and the company is candid about that in its own numbers: RMB 200 to 300 million of robot orders in 2026 is a foothold, not a business.
For context on how far the domestic base has come, China installed 354,000 industrial robots in 2025, about 59 per cent of the global total. The components feeding that installed base are now largely domestic. The question the humanoid cycle is asking is whether the same substitution can happen one level up, in the actuator that makes a joint move. Leadshine’s order book is one company’s answer.
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. The full original (in Chinese) is at https://robot.ofweek.com/2026-10/ART-8321201-12003-30709252.html.