Image: ASUS — Gallop (official product image)
TL;DR — ASUS Gallop humanoid robot explained: ASUS unveiled Gallop, its first humanoid robot, on 13 November 2026. Gallop is learning a factory job: packing 15 motherboard accessories in about 20 seconds a board, judged by the same 99.5% success standard as a person. It runs on one Nvidia Jetson AGX Thor computer, with pilots in 2027 and sales targeted for late 2028.
Key Highlights
- ASUS Gallop is ASUS's first humanoid robot, unveiled on 13 November 2026 in an ASUS Pressroom post, after co-CEO S.Y. Hsu introduced it on CNBC.
- Its first job is a real one: packing 15 motherboard accessories at a workstation in an ASUS group electronics manufacturing plant, at about 20 seconds per board.
- ASUS's bar is blunt: 99.5% success within 20 seconds — the same standard it holds a human worker to. That is a target, not a result achieved so far.
- Gallop runs on a single Nvidia Jetson AGX Thor computer inside the robot, with a four-layer AI stack and a body and hand designed in-house.
- Customer pilots are planned for 2027, with market availability targeted for the second half of 2028. No price, unit cost or production volume has been disclosed.
What happened?
ASUS, the Taiwanese company most readers know for motherboards, laptops and ROG gaming gear, has entered the humanoid robot race. On 13 November 2026 it published a Pressroom post, "Factory-Proven Physical AI: Why ASUS Is Building a Humanoid Robot", introducing Gallop — its first humanoid — shortly after co-CEO S.Y. Hsu presented the robot on CNBC.
Gallop comes from the Humanoid Robotics business unit inside ASUS's Physical AI Solutions Business Group, and the programme started in early 2026, so the robot has been in development for less than a year. What makes the announcement stand out from the usual humanoid reveal is what ASUS chose to show first: not a backflip or a dance routine, but a robot learning a packing station on a real production line, judged by the same pass rate and cycle time as the person who does the job today.
ASUS's own framing is that videos of humanoids are everywhere, but a factory manager asks a different question: can it do this station, for a full shift, to my quality standard? Gallop is ASUS's attempt to answer that question at home first, in its own factories, before selling it to anyone else.
Gallop's first job: 15 parts, 20 seconds, one hard test
Gallop's first assignment is packing motherboard accessories at an ASUS group electronics manufacturing services (EMS) plant — a factory that builds electronics for other companies. The station holds 15 small parts, many of them in thin, clear plastic bags, and takes about 20 seconds per board. ASUS's target is 99.5% success within 20 seconds, matching the standard it sets for a person doing the same job.
ASUS picked this station deliberately because it is hard. Small, thin and soft parts in slippery bags are exactly what a simple gripper cannot handle, and the station tests the rule ASUS judges itself by: a robot has to take on most of a job. If a person still has to stand beside it the whole time, ASUS says, nobody gains anything.
Two honest caveats matter here. First, 99.5% within 20 seconds is a target ASUS has set, not a measured result — the company has not disclosed how often Gallop succeeds at the station today. Second, the robot's in-house dexterous hand has not been integrated yet; ASUS says it will bring its own hand into the robot over the coming year. What ASUS says does work today is the full loop: a person shows the task, the robot learns it, and the company tests it on the line.
| Detail | Gallop at a glance |
|---|---|
| What it is | ASUS's first humanoid robot, from the Humanoid Robotics business unit of the Physical AI Solutions Business Group |
| First job | Packing 15 motherboard accessories per board at an ASUS group EMS factory |
| Speed target | About 20 seconds per board, at 99.5% success — the same standard as a human worker (target, not a measured result) |
| Compute | One Nvidia Jetson AGX Thor computer inside the robot, standard robotics software |
| Form today | Fixed base, two arms, pick-and-place work; wheeled version planned |
| Timeline | Customer pilots in 2027; shown to customers at Computex 2027; market availability targeted for the second half of 2028 |
| Price | Not disclosed — no price, unit cost or production volume announced |
ASUS Gallop humanoid robot explained: how it learns from factory workers
Teaching Gallop works much like training a new colleague, in five steps. First, ASUS starts from leading open AI models that already understand how hands and objects behave, rather than building its own giant model. Second, experienced workers wear a small head camera — with their consent — so Gallop can learn how the job is really done. Third, simulation generates variations of each example, so small changes on the line do not surprise the robot. Fourth, an operator in a VR headset and gloves guides the robot through the task, on the same parts and with the same cameras as the real station. Fifth, once deployed, what the robot records feeds back into training.
ASUS says its edge is not more data but the right data: a real station, real parts and real quality rules. Its workers are the teachers, and their know-how is what Gallop is built on. That is also why ASUS thinks it has an advantage over most robotics startups: it owns the electronics factories where the robot is tested, on real parts under real quality rules, instead of having to rent that environment.
What's inside Gallop: one Nvidia Thor computer and four AI layers
Inside, Gallop works in four layers. A "planner" looks at the scene, understands the job and breaks it into steps. A "supervisor" — a second AI — checks every step and re-plans when something goes wrong. A skills layer handles reaching, grasping and placing. And a control layer moves the arms and hands precisely, and stops instantly if anything is wrong.
The layers talk to each other through fixed interfaces, so when a better AI model arrives — and new models now arrive every few months — ASUS can swap it in without rebuilding the robot. All of this runs on a single Nvidia Jetson AGX Thor computer inside the robot, using standard robotics software. ASUS designs the robot body and the hand itself; the hand is the hardest part of any humanoid and is still scarce worldwide, so ASUS builds its own in-house, with a sense of touch. Motors and other standard components come from established partners.
Image: ASUS — close-up of Gallop's hand and vision sensor at the packing station (official product image)
Today Gallop works from a fixed base with two arms, and its strength is pick-and-place work at a station. Over the coming year ASUS plans to bring its own dexterous hand into the robot, put Gallop on wheels, and show it to customers at Computex 2027. Electronics manufacturing comes first, because it is the industry ASUS knows best — and because many jobs share the same pick-and-place skills, what Gallop learns in one place can carry over to the next. ASUS sees hospitals and service sites as later markets.
Why it matters for India readers
India will not be buying a Gallop any time soon — there is no India price, no India launch date, and market availability is targeted for the second half of 2028 at the earliest. But the story still lands close to home. ASUS is a familiar brand in India, from the motherboards in countless Indian-built PCs to its ROG laptops, and India is now one of the world's fastest-growing electronics manufacturing hubs, with factories assembling phones, laptops and components at rising volumes.
The trend Gallop represents is the one those factories will have to judge: humanoids that slot into workstations built for people, rather than automation that demands the factory be rebuilt around it. If ASUS can prove a robot holds a 99.5% pass rate at human speed on a real line, that becomes the benchmark every electronics manufacturer — in India included — will measure rivals against. And the way Gallop learns, from the workers already on the line, is the version of factory automation that treats their experience as the training data, not as something to discard.
The wider race is moving at the same pace. GXO's plan to put 1,000 Apptronik Apollo humanoids into warehouses targets a late-2027 rollout, and J.P. Morgan expects humanoid production to reach 400,000 units a year by 2030. Gallop is ASUS's entry into that same race, from the factory floor rather than the warehouse. Nvidia's GR00T 2 robot model, which learns dexterous skills from human video, shows the same learning-from-people idea spreading across the industry — and Gallop's brain runs on Nvidia's Thor hardware.
Should you care?
If you follow robotics or manufacturing tech, yes — watch the 2027 pilots. Gallop is one of the few humanoids being judged in public against a human pass rate on a real line, and the pilot results will show whether that bar is reachable. If you were hoping to buy one, wait. There is no price and no sale before the second half of 2028, and the wheeled version and the in-house hand are still to come. The honest read: this is a credible, carefully scoped start — one station, one factory, one clear target — not a robot that can do everything, and ASUS is saying so itself.
Price, availability and alternatives
ASUS has disclosed no price, unit cost or production volume for Gallop. The timeline it has given is: customer pilots planned for 2027, a showing to customers at Computex 2027, and market availability targeted for the second half of 2028. Those are targets, and robotics schedules have slipped before — treat the dates as ASUS's plan, not a promise.
The competitive field is crowded. Apptronik's Apollo is heading for warehouses under GXO's robotics-as-a-service rollout, Tesla's Optimus and Figure's humanoids are chasing general factory and household work, and Nvidia's robot models are supplying brains to several of these machines at once. Gallop's differentiator is narrower and more concrete than most: a PC maker testing its own robot in its own factories, on one station, against one human standard. If that works, the same pick-and-place skills transfer to the next station — and eventually, ASUS hopes, to hospitals and service sites.
FAQs
What is ASUS Gallop? Gallop is ASUS's first humanoid robot, unveiled on 13 November 2026. It is a two-armed robot learning to pack motherboard accessories at a workstation in an ASUS group electronics factory, running on a single Nvidia Jetson AGX Thor computer. It is built by the Humanoid Robotics business unit of ASUS's Physical AI Solutions Business Group.
When will ASUS Gallop be available? Customer pilots are planned for 2027, ASUS plans to show Gallop to customers at Computex 2027, and market availability is targeted for the second half of 2028. Those are ASUS's stated targets, not confirmed launch dates.
How much will ASUS Gallop cost? ASUS has not disclosed a price, unit cost or production volume. No pricing information was included in the 13 November announcement, and the robot is not for sale yet.
Will ASUS Gallop replace factory workers? ASUS says no. Gallop is designed for repetitive, physically demanding stations that are hardest to staff, and the company says the workers who teach it move into setting up, supervising and improving the robots — with their experience becoming more valuable, not less.
Takeaway
ASUS is not selling a robot that can do everything — it is proving one robot can do one factory job at human speed, every day. If the 2027 pilots hold that 99.5% bar, Gallop becomes the template for how humanoids enter real factories. If they don't, ASUS will have told us so itself: it has promised to keep sharing its progress against that standard.
The next date that matters is Computex 2027, when ASUS plans to show Gallop to customers — until then, our explainer on GXO's 1,000-robot Apptronik rollout shows how the warehouse side of the same humanoid race is moving.
Sources
ASUS Pressroom — Factory-Proven Physical AI: Why ASUS Is Building a Humanoid Robot (13 November 2026)
Unite.AI — ASUS Sets Human-Pace Packing Target for Gallop Humanoid Robot (13 November 2026)
Interesting Engineering — ASUS builds humanoid robot to pack tiny parts in electronics factories (13 November 2026)
Comments
Post a Comment