ASUS Targets 99.5% Packing Success for Gallop Humanoid Robot

ASUS wants its first humanoid robot, Gallop, to pack 15 motherboard accessories with a 99.5% success rate within 20 seconds per board. The company detailed those targets on November 13, 2026 — and they are targets, not achieved results.

Gallop is learning the packing job at one of the group's own electronics factories. Many of the parts sit in thin, clear bags, which makes the workstation a hard assignment for simple grippers. ASUS picked it deliberately: the test is whether the robot can handle most of a job without a person standing beside it the whole time. The company applies the same speed-and-success standard to human workers.

The full loop now works end to end. A person demonstrates the task, the robot learns it, and ASUS tests it on the line. The Humanoid Robotics business unit, inside the Physical AI Solutions Business Group, runs the program, which began in early 2026.

ASUS's argument for the humanoid shape is simple economics: arms and dexterous hands can use the workstations, tools and trays already designed for people, so a factory doesn't need to be rebuilt around the robot. The company concedes not every task needs a humanoid.

How does Gallop learn?

ASUS starts with open AI models rather than building a giant model in-house. Experienced workers, with their consent, wear head-mounted cameras so their demonstrations capture how the job is actually performed. Simulation then generates variations of those examples.

An operator wearing a VR headset and gloves guides Gallop through practice with the same parts and cameras used at the real station. Recordings made after deployment feed back into training.

ASUS describes its edge as the right data — from real stations, real parts and real quality rules — with its own electronics manufacturing services factories supplying a test environment that most robotics companies must rent. That's the takeaway: ASUS owns its proving ground.

What's inside the robot?

Gallop's software stack has four fixed layers:

• A planner interprets the scene and splits the job into steps
• An AI supervisor checks each step and replans after a problem
• A skills layer handles reaching, grasping and placing
• A control layer moves the arms and hands precisely and stops them if something is wrong

The fixed interfaces between layers are meant to let ASUS swap AI models without rebuilding the robot. Everything runs on one NVIDIA Jetson AGX Thor computer inside Gallop, using standard robotics software.

ASUS designs the body and a touch-sensing dexterous hand in-house; motors and other standard components come from established partners. The current robot uses a fixed base and two arms for stationary pick-and-place work. The in-house hand has not yet been integrated.

When can customers buy one?

The roadmap is specific. ASUS plans to integrate the hand, add wheels, and show Gallop to customers at Computex 2027. Customer pilots are also planned for 2027. Market availability is targeted for the second half of 2028.

Electronics manufacturing is the first intended market. ASUS expects pick-and-place skills learned in one job to transfer to others, opening possible later uses in hospitals and service sites.

The company positions the robot for repetitive, physically demanding stations that are hard to staff — not as a replacement for workers. Employees who teach Gallop would move into setting up, supervising and improving robots.

The 2027 pilots carry the real weight. They will test whether learning from skilled workers and station-specific data can hold factory quality and pace through daily operation. ASUS has made that consistency the threshold for broader use: the working training loop is the starting point, and the pilots must prove it delivers repeatable results under everyday factory conditions.