1X Starts Full-Scale NEO Production in California, Targets 10,000 Home Robots a Year

1X says its new 58,000-square-foot NEO factory in Hayward, California is now in full-scale production, with capacity to build 10,000 home humanoid robots per year and a roadmap toward 100,000-plus annual units by the end of 2027. Those are unusually specific manufacturing claims for a category that still spends most of its time on polished demo clips.

That is why this announcement matters. NEO is not being positioned as another lab prototype or a factory pilot for a single industrial customer. It is being pitched as a consumer home robot, with early access priced at $20,000, subscription pricing at $499 per month, and first customer shipments still promised for 2026. In humanoid robotics, the hard part is no longer just making a robot walk, grasp, or hold a conversation. The hard part is building thousands of reliable machines, supporting them, and improving them fast enough once they land in real homes.

AI-generated image

Editorial illustration of a high-volume robotics production floor focused on motors, batteries, and final assembly systems. Source: biped.news

How 1X Plans to Turn a Home Robot Into a Manufactured Product

Most humanoid announcements still revolve around capability, funding, or partnerships. 1X is trying to push the conversation one step later in the stack, into manufacturing. The company says the Hayward site is America’s first vertically integrated high-volume humanoid robot factory, and the phrase that matters there is not really "humanoid." It is vertically integrated. According to the company, it is building motors, batteries, transmission systems, sensors, soft goods, and structural components in-house instead of leaning heavily on outside subsystem vendors.

That approach is expensive and operationally messy, but there is a reason startups keep chasing it. If your robot is meant for the home, field reliability and iteration speed matter more than slide-deck elegance. A vertically integrated setup lets 1X watch a failure in testing, revise a part, push the change into production, and feed the result back into software and validation without waiting on three different suppliers. The company says the factory came together in roughly three months after permits and is already fully operational with more than 200 employees. That is fast by any hard-tech standard.

1X says its October NEO preorder launch booked out the next year of planned production capacity in five days, which it quantifies as 10,000 NEOs. That is a company-reported demand signal, not the same thing as 10,000 completed customer deliveries. The factory evidence is stronger on what exists today: 1X's April 30 manufacturing update says the 58,000-square-foot Hayward site is operational and has capacity for up to 10,000 NEOs per year.

Under the Hood: The Manufacturing Stack Behind NEO

The company’s factory materials offer more technical detail than most consumer robot launches. 1X says it has already manufactured 17,000 Revo2 motors for NEO’s tendon-driven actuation system. That matters because the robot’s core pitch is not brute industrial power. It is safe, quiet, human-scale motion. NEO is designed around tendon drives, soft outer materials, and low-noise operation. 1X says the system runs at about 22 decibels, quieter than a modern refrigerator, which is exactly the kind of metric a home robot has to win on if it expects to stay in the room with people all day.

The battery stack is also unusually specific. 1X says its pack manufacturing line can produce 10,000 batteries per year, and that the company’s battery management system samples at 100 hertz, roughly 100 times faster than a typical commercial pack, according to its own materials. The point is not just battery life, which outside summaries place at roughly four hours. The point is thermal control, state estimation, and fast charging under a home-use duty cycle where reliability matters more than a headline top speed.

Then there is compute. Every production NEO is slated to ship with NVIDIA Jetson Thor as the core of the company’s NEO Cortex. 1X says that stack handles real-time perception, reasoning, and decision-making on the robot itself, with Isaac Sim and Isaac Lab used for simulation and reinforcement learning. This is now the standard physical AI playbook: do as much inference on the machine as possible, use simulation to compress training cycles, and keep cloud dependence off the critical path for safety-sensitive tasks.

AI-generated image

Editorial illustration highlighting the component layer behind a home humanoid, including tendon-drive hardware, electronics, and test instrumentation. Source: biped.news

1X: The Full Picture

1X sits in an unusual position inside the humanoid market. The company is backed by OpenAI and last raised about $100 million in a January 2024 Series B led by EQT Ventures, bringing total disclosed funding to about $123.5 million according to recent reporting. That is meaningful capital, but it is still small next to the war chests being assembled around Figure, Tesla, and some of the larger China-based robot makers. 1X does not win by spending the most. It wins by finding a narrower product thesis and executing faster.

Its thesis is clear enough: a soft, quiet, home-first humanoid can open a market that factory-only robots do not touch. Public spec summaries place NEO at about 170 centimeters tall, 30 kilograms, with 22-degree-of-freedom hands, around 70 kilograms of lift strength, and roughly 25 kilograms of carry capacity. Those figures are not the same as saying the robot is ready for unstructured domestic work at scale. But they do show that 1X is optimizing for a different experience from the heavier industrial humanoids that dominate most headlines.

The company also seems more willing than most to talk plainly about operations. Its public materials dwell on hiring, training centers, metrology, fit-and-finish checks, warehousing, and internal home testing. That may sound boring compared with viral demo clips, but boring is healthy in robotics. You want a team obsessed with weld rates, calibration drift, camera modules, and field returns. Those details decide whether a robot becomes a product or remains a press release.

What This Means for the Consumer Humanoid Market

The home humanoid market has had no shortage of big claims. What it has lacked is a believable manufacturing story. That is why 1X’s factory matters beyond the company itself. If NEO can move from preorders to real deliveries, the market gets a badly needed data point on what a consumer robot rollout actually looks like. Not a concept video, not a pilot inside the parent company’s own buildings, but units shipped to outside users with expectations, complaints, and support tickets.

It also puts pressure on competitors. Figure has been talking more about autonomy and production ramp. Unitree is pushing price compression hard. Tesla still commands the most public attention, but its path runs through a very different manufacturing machine and a much broader corporate agenda. 1X is carving out a lane where softness, quietness, and domestic fit matter more than warehouse theatrics or celebrity founder energy. That could be smart if the first durable consumer demand ends up clustering around chores, elder assistance, and simple telepresence rather than full general labor.

AI-generated image

Editorial illustration of the operations side of consumer robot manufacturing, including packaging, quality checks, and warehouse flow. Source: biped.news

There are still obvious risks. A preorder book and stated factory capacity are not the same thing as repeatable revenue or units delivered. 1X's current order page still says U.S. deliveries start in 2026, while the company's factory update says current units are being prioritized for internal development and home testing as manufacturing ramps. A robot that works in internal testing can still struggle in ordinary homes where clutter, pets, lighting, narrow hallways, and uneven floors break assumptions. The next evidence threshold is customer delivery, useful task completion, intervention burden, reliability, and support cost.

What's Coming Next

The next twelve months will test whether 1X’s factory announcement translates into something the market can verify. The first milestone is simple: do customer shipments begin in 2026 as promised. The second is harder: once those units leave Hayward, how many tasks can they actually complete in ordinary homes without remote rescue or extensive setup. The third is economic. If 1X really can scale toward 100,000 annual units by the end of 2027, then component cost, service coverage, and software update cadence become the story.

There is also a strategic question hanging over the whole sector. Will the first winning consumer robot be a highly general humanoid, or a carefully constrained household assistant that gets better through teleoperation, software updates, and narrow task expansion. 1X appears to be betting on the second path. That is less romantic than the sci-fi version of the future, but it is probably the more plausible one.

Frequently Asked Questions

The 12-Month Outlook

If 1X hits even the first part of its shipping promise, this will count as one of the most important consumer robotics milestones of 2026. Not because NEO is guaranteed to dominate the category, but because a real shipment cycle would give the market a way to judge the home humanoid thesis on service quality, reliability, and retention instead of hope.

For now, the most credible reading is simple. 1X has moved the conversation from what a home humanoid might do to whether one can be built, tested, and delivered at meaningful scale. That is a harder question, but it is finally the right one.

If the company ships on time and the robots hold up in the field, 1X will have done more than launch a product. It will have given the whole sector a new benchmark for what "real" progress looks like.