# WhichBot — Full Content for AI Engines

Independent buyer's guide to home robots and construction & building robots.
Data updated 2026-07-04. Figures marked * are estimates from public sources.
Not affiliated with any manufacturer. Contact: support@whichbot.net

============================================================
# ROBOT PROFILES

## Tesla Optimus Gen 3 (Home & personal robots)
URL: https://whichbot.net/robots/tesla-optimus
Price: $25,000*. Elon Musk has repeatedly targeted a $20,000–$30,000 price at volume; no official consumer price has been announced.
Availability: Waitlist — Tesla has converted Model S/X production lines to build Optimus. No confirmed consumer ship date; industrial/internal deployments first.
Specs: Height: 5′8″ / 173 cm; Weight: ~57 kg; Payload: 20 kg; Runtime: ~8 hrs; Dexterity: 28+ DOF (11 per hand)
WhichBot score: 8.1/10. Best for: Buyers who want to bet on the biggest ecosystem and can wait for a consumer launch.
Strengths: Manufacturing scale nobody else can match; Tesla's AI/data flywheel from its vehicle fleet; Likely the most aggressive consumer price at volume
Weaknesses: No confirmed consumer release date — timelines have slipped before; Unproven in real homes; Early units go to Tesla factories, not buyers
The case for Optimus has never been about the robot you can see today — it's about the factory behind it. Tesla is the only company in this index that already mass-produces complex electromechanical machines by the million, and it has now converted Model S/X production lines to build Optimus. That matters more than any demo: humanoids are currently expensive because they're hand-built in small numbers, and Tesla's entire thesis is that automotive-scale manufacturing collapses that cost. If the company lands anywhere near its stated $20,000–$30,000 target, it doesn't just join the market — it reprices it.
The honest counterweight is history. Tesla timelines are directional, not contractual: Full Self-Driving, the Cybertruck, and the Semi all arrived years behind their announcements. Optimus today works in Tesla's own facilities, not in anyone's home, and no consumer reservation system exists. Buyers who wait for Optimus are making a rational bet on the biggest ecosystem in the category — but they should size that bet knowing the ship date is unknowable, and that 1X is putting robots into actual houses while Tesla builds factory capacity.
Our read: Optimus is the robot to wait for if you have no urgent need and want maximum capability per dollar when the consumer market truly opens. Join the information waitlist, watch for a reservation system (Tesla's vehicle playbook suggests one is coming), and let the early-adopter tax fall on someone else.

## 1X Technologies NEO Gamma (Home & personal robots)
URL: https://whichbot.net/robots/1x-neo
Price: $20,000*. 1X has discussed a price around $20,000 or a monthly subscription; final consumer pricing may differ.
Availability: Home pilot — In limited in-home pilot programs now — the only full-size humanoid actually living in consumer houses today.
Specs: Height: 5′6″ / 168 cm; Weight: ~30 kg; Payload: 15 kg; Runtime: ~4 hrs; Dexterity: Soft-actuated, human-level range
WhichBot score: 8.4/10. Best for: Households that want a robot around people, kids, and pets as early as possible.
Strengths: Actually deployed in real homes right now; Soft-body, quiet design made for living spaces; Safety-first engineering (low mass, compliant joints)
Weaknesses: Lower payload than industrial rivals; Pilot access is limited and regional; Relies partly on teleoperation for hard tasks today
NEO is the only entry in this index answering the question every consumer actually asks: what is it like to have a humanoid living in my house right now? Through 1X's expanding home pilot, NEO units are doing exactly that — fetching, carrying, tidying, and monitoring in real family homes. Everything about the machine is built around that setting: at roughly 30 kg it's less than half the weight of industrial rivals, its soft-body construction turns collisions into bumps rather than injuries, and it runs quietly enough to share a living room with a sleeping baby.
The trade-offs are real and worth stating plainly. NEO's payload and raw dexterity trail warehouse machines like Figure 03 — a deliberate engineering choice, since compliant joints that are safe around children can't also be forklift-strong. And like every home humanoid in 2026, NEO leans on remote teleoperation for tasks its AI can't yet handle autonomously; 1X is unusually transparent that assisted operation doubles as training data. Pilot access is also limited and regional, so 'in homes now' doesn't yet mean 'in your home next week.'
Our read: if the goal is a robot around your people as early as possible — especially with kids, pets, or an older parent in the picture — NEO is the pick, and it isn't close. Safety-first design is the feature that matters most in a home, and NEO is the only machine proving it in production households rather than promising it on a keynote stage.

## Unitree G1 (Home & personal robots)
URL: https://whichbot.net/robots/unitree-g1
Price: $16,000. Unitree publishes list pricing; configuration and import costs vary.
Availability: Buy today — Orderable today. Unitree also became the first humanoid-robot IPO in 2026, signalling scale-up ahead.
Specs: Height: 4′3″ / 130 cm; Weight: ~35 kg; Payload: 3 kg; Runtime: ~2 hrs; Dexterity: 23–43 DOF depending on config
WhichBot score: 7.2/10. Best for: Tinkerers, developers, and early adopters who want hardware in hand now.
Strengths: You can actually buy it, today; Lowest price of any real humanoid; Huge developer/hobbyist community
Weaknesses: Small stature and 3 kg payload limit chores; Developer platform, not a housekeeper; Software skills required to get value
The G1's superpower is the least glamorous one in robotics: you can actually buy it. Sixteen thousand dollars, a real order page, a shipping date — while most of the industry sells futures, Unitree sells hardware. The company's 2026 IPO (the industry's first) tells you the strategy is working, and its scale gives the whole market's price curve a persistent downward pull. As a piece of engineering, the G1 is remarkable for the money: a genuinely dynamic biped with up to 43 degrees of freedom and a developer ecosystem that has produced everything from boxing demos to household experiments.
Know what you're buying, though. At 4′3″ and a 3 kg payload, the G1 is not going to carry your laundry basket or reach your kitchen counters — it's a platform, not a housekeeper. Getting real utility out of it means software work, either your own or the community's. Budget for import considerations, and treat capability videos the way you'd treat any developer-platform demo: as what's possible, not what's included in the box.
Our read: for tinkerers, developers, researchers, and content creators, the G1 is the best value in robotics — frontier hardware at a hobbyist-accessible price. For someone who wants a robot that just helps around the house with zero technical effort, it will frustrate; that buyer should look at NEO or wait for the consumer wave.

## Figure AI Figure 03 (Home & personal robots)
URL: https://whichbot.net/robots/figure-03
Price: Not for sale. Figure sells to commercial customers only; no consumer pricing exists.
Availability: Commercial only — Deployed in logistics at scale — hundreds of thousands of packages processed with zero hardware failures reported at its Sunnyvale facility.
Specs: Height: 5′6″ / 168 cm; Weight: ~60 kg; Payload: 20 kg; Runtime: ~5 hrs; Dexterity: High-dexterity hands (Helix AI)
WhichBot score: 8.7/10. Best for: Businesses — and consumers watching where home capability will come from next.
Strengths: Best-in-class manipulation via the Helix AI stack; Proven at scale in commercial logistics; Autonomous battery-swap operation
Weaknesses: Consumers cannot buy it; No announced home product or price; Enterprise sales cycle only
On pure capability, Figure 03 may be the best humanoid on Earth. Its Helix AI stack drives what are arguably the most dexterous hands in the industry, and the receipts are commercial, not theatrical: hundreds of thousands of packages processed in logistics deployment with zero reported hardware failures, running autonomous battery swaps to work around the clock. When people in the industry talk about the state of the art in real-world manipulation, this is the machine they mean.
The catch for readers of this site is simple: you cannot buy one at any price. Figure sells to enterprises, has announced no home product and no consumer pricing, and its multi-billion-dollar valuation is built on commercial contracts. We track it anyway, for a reason that matters to consumers: capability flows downhill. The manipulation skills being proven in Figure's warehouse deployments today are a preview of what home robots inherit in the next few years — either from Figure itself entering the consumer market, or from competitors racing to match it.
Our read: for businesses with high-volume manipulation work, Figure 03 is the capability benchmark against which everything else is judged. For consumers, it's the most important robot you can't buy — watch it to understand where your future home robot's skills are coming from.

## Apptronik Apollo (Home & personal robots)
URL: https://whichbot.net/robots/apptronik-apollo
Price: Lease only. Apptronik offers commercial leasing / robots-as-a-service; no consumer price.
Availability: Commercial only — Commercial deployments in warehousing and manufacturing, including work with major logistics partners.
Specs: Height: 5′8″ / 173 cm; Weight: ~73 kg; Payload: 25 kg; Runtime: ~4 hrs (swappable); Dexterity: Modular, swappable batteries
WhichBot score: 7.9/10. Best for: Warehouse and manufacturing operators — not home buyers.
Strengths: Highest payload of any humanoid in this index (25 kg); Hot-swappable batteries = continuous shifts; Designed for safety in human workspaces
Weaknesses: Not available to consumers; Utilitarian design, industrial focus; Lease-only economics
Apollo is what happens when NASA-heritage engineers design a humanoid for work rather than wonder. Its 25 kg payload is the highest in this index, its batteries hot-swap so the machine never stops for a charge, and everything about its friendly-but-plain design says warehouse shift, not living room. Apptronik's partnerships with major logistics and manufacturing operators have moved Apollo into real commercial deployments where those choices pay off daily.
The limits are the same ones that define its category: Apollo is lease-only, enterprise-only, and utilitarian by intent. There is no consumer version, no retail price, and no pretense of one. Its economics make sense for operators running shifts, where a tireless 25-kg-payload worker with swappable batteries competes directly against labor shortage math — not for households.
Our read: for warehouse and manufacturing operators whose bottleneck is repetitive heavy handling, Apollo is a serious, proven option and arguably the purest work machine in the humanoid market. For everyone else, it's a demonstration of how quickly 'robot coworker' went from concept to line item.

## Agility Robotics Digit (Home & personal robots)
URL: https://whichbot.net/robots/agility-digit
Price: RaaS pricing. Agility sells via robots-as-a-service contracts; pricing is quoted per deployment.
Availability: Commercial only — The longest continuous real-world humanoid deployment record, running in third-party logistics facilities for years.
Specs: Height: 5′9″ / 175 cm; Weight: ~65 kg; Payload: 16 kg; Runtime: ~4 hrs; Dexterity: Bird-legged biped, tote-handling arms
WhichBot score: 7.5/10. Best for: Logistics operators who value a proven track record over frontier specs.
Strengths: Years of real deployment data; Purpose-built reliability for logistics; RaaS model lowers adoption risk
Weaknesses: Not a consumer product; Specialised for tote/box handling; Distinctive design isn't home-oriented
Digit's claim to fame is time. While competitors were filming demos, Agility's bird-legged biped was logging years of continuous deployment in third-party logistics facilities — the longest real-world track record of any humanoid, full stop. That history shows up in the design: Digit isn't trying to look human, it's trying to move totes reliably for thousands of hours, and its distinctive reverse-knee legs are optimized for exactly the crouch-lift-walk cycle warehouse work demands.
Agility sells through robots-as-a-service contracts, which lowers the risk of adoption for operators but closes the door to consumers entirely. Digit also isn't chasing general-purpose dexterity the way Figure is — it's a specialist, brilliant at tote and box handling and uninterested in your kitchen. That focus is a strength for its customers and the reason it will never appear on a home-robot shortlist.
Our read: Digit is the existence proof of the entire category — the robot that demonstrated humanoids can do economically useful work year after year, not just on stage. Logistics operators who value a track record over frontier specs should have it on the shortlist; consumers should appreciate it from a distance.

## Skyline Robotics Ozmo (Construction & building robots)
URL: https://whichbot.net/robots/skyline-ozmo
Price: RaaS / quote. Ozmo is offered as a robot-as-a-service platform; pricing depends on facade size and cleaning schedule.
Availability: Service contract — In full-time commercial deployment — including a 45-story Class A tower at 1133 Avenue of the Americas in Manhattan — with expansion underway in London and patents secured for Japan and Singapore.
Specs: Task: High-rise window cleaning; Speed: ~3× faster than human crews; Platform: Kuka arm on existing BMU cradle; Sensing: LiDAR + force sensors + vision; Crew: 1 rooftop operator (vs 3-person crew)
WhichBot score: 8.6/10. Best for: Class A commercial towers and facade-maintenance contractors who want faster, safer window cleaning without retrofitting the building.
Strengths: Actually deployed full-time on major NYC towers; Works with the building's existing cradle/BMU — no retrofit; Force sensors adapt pressure to fragile glass; handles wind gusts
Weaknesses: Service model only — you can't buy the hardware; Requires a certified operator on the roof (regulatory); Best suited to buildings that already have a maintenance unit
Ozmo is the rare robot whose pitch requires no imagination: look up in Midtown Manhattan and it's already working. In full-time deployment on a 45-story Class A tower, Skyline's system mounts a Kuka robotic arm on the building's existing window-washing cradle and cleans roughly three times faster than a human crew — with one certified operator supervising from the roof instead of a multi-person team suspended on the wall. LiDAR maps the facade in real time, force sensors modulate pressure for fragile glass, and the AI holds the system steady through wind gusts.
The genius of the design is what it doesn't require: no building retrofit. Ozmo rides the maintenance unit (BMU) that high-rises already have, which turns adoption from a construction project into a service contract. That's also its constraint — buildings without a roof rig, or with highly ornamental facades, remain human territory. Skyline sells outcomes, not hardware, priced per building by facade size and schedule, and regulations still require the human rooftop operator (a feature, honestly, in a young category).
Our read: for Class A towers with existing BMUs and recurring cleaning contracts, Ozmo has crossed from innovation-committee curiosity to straightforward procurement decision — get a service quote, compare it to your current cleaning contract, and let the math decide. The labor context (an aging window-cleaning workforce with few young entrants) means this comparison only tilts one direction over time.

## Boston Dynamics Spot (Construction & building robots)
URL: https://whichbot.net/robots/boston-dynamics-spot
Price: From $74,500. Base platform list price; construction payloads (laser scanners, cameras, docks) are additional.
Availability: Buy today — Purchasable today and widely deployed on major construction projects for progress scans, QA documentation, and safety walks — often paired with laser-scanning payloads.
Specs: Task: Site inspection, scanning & monitoring; Speed: Autonomous repeat site walks; Platform: Quadruped, stairs & rough terrain; Sensing: 360° perception + payload sensors; Crew: Autonomous missions after setup
WhichBot score: 8.2/10. Best for: General contractors who want automated progress documentation and QA scanning across large or repeat projects.
Strengths: You can buy it today with a published price; Handles stairs, rubble, and real job-site terrain; Huge payload ecosystem: laser scanners, 360° cameras, docks
Weaknesses: It inspects and documents — it doesn't build anything; Total cost climbs fast with payloads and software; Needs mission setup and a data workflow to earn its keep
Spot is the construction robot with the least mystery and the most receipts: a published price (from $74,500), an open sales channel, and years of deployments walking real job sites. Its job is discipline, not construction — the scheduled site walk that humans skip when things get busy. Spot performs it identically every time: same route, same scan positions, same photo angles, producing the consistent progress documentation and QA data that keeps disputes short and rework visible early. Stairs, rubble, and half-finished floors that stop wheeled robots are exactly what its legs are for.
The full cost story deserves honesty: the base platform is the start, not the end. A working construction setup typically adds a laser scanner or 360° camera payload, mission software, and someone's time to design the routes and plumb the data into your project systems — a loaded rig can run double the sticker. Spot also builds nothing; its ROI lives entirely in what documentation catches, which is why it repays fastest (often under six months) on large, complex, or dispute-prone projects.
Our read: for general contractors running big or repeat projects, Spot is the most proven, lowest-drama entry point into construction robotics — a purchase decision, not a leap of faith. Just budget for the payload ecosystem, not the headline price.

## FBR Hadrian X (Construction & building robots)
URL: https://whichbot.net/robots/fbr-hadrian-x
Price: Walls-as-a-Service. FBR offers Hadrian X through a Walls-as-a-Service program — contractors hire the system per project rather than buying the machine.
Availability: Service contract — Moving into full commercial availability as of mid-2026 after completed pilots in the US and Mexico, offered via a Walls-as-a-Service model.
Specs: Task: Robotic bricklaying / blocklaying; Speed: Up to ~360 large blocks/hr; Platform: Truck-mounted 30 m telescoping boom; Sensing: Laser-guided dynamic stabilization; Crew: Small supervising team + block loading
WhichBot score: 8/10. Best for: Homebuilders and low-rise commercial contractors facing masonry labor shortages who want structural walls dramatically faster.
Strengths: Can complete the walls of a house in about a day; Builds directly from the digital model — millimetre accuracy; Service model means no capital purchase to start
Weaknesses: Requires custom large-format blocks and adhesive system; Suited to low-rise structures, not high-rise cores; Availability is via select partners, not open booking everywhere
Hadrian X answers construction's most brutal labor equation — there are fewer licensed masons in America than at any point since 1950 — with a machine that lays walls from a truck. Its 30-meter telescoping boom places large-format blocks (roughly twice standard brick size) directly from the digital model at up to ~360 blocks per hour, using laser-guided dynamic stabilization to stay millimetre-accurate even in wind. It cuts blocks on board, leaves openings for windows and doors, and bonds everything with a rapid construction adhesive that FBR states outperforms traditional mortar. The demonstration that gets attention: walls of a house in about a day.
The 2026 milestone is commercial, not technical: after completed pilots in the US and Mexico, Hadrian X has moved into full availability through a Walls-as-a-Service model — contractors hire finished walls per project rather than buying the machine. That structure removes the capital-expenditure barrier but places access through FBR's partner network, and the system's sweet spot is straight production walls on low-rise structures. Ornamental masonry, repairs, and high-rise cores remain human trades.
Our read: for homebuilders and low-rise commercial contractors whose schedules bleed at the masonry line item, Hadrian X is now a quotable service — price a project against your current masonry sub and let the schedule compression do the arguing. It's the most ambitious machine in this index that you can actually book.

## Hilti Jaibot (Construction & building robots)
URL: https://whichbot.net/robots/hilti-jaibot
Price: Quote / lease. Hilti offers Jaibot through its fleet-management and leasing programs; pricing is quoted per project or contract.
Availability: Lease / fleet — In field use since 2020 with major contractors worldwide (Skanska, Bouygues and others) — arguably the most proven construction robot on the market.
Specs: Task: Overhead drilling for MEP installs; Speed: Weeks of ceiling work → days; Platform: Tracked mobile base, BIM-guided; Sensing: Total-station positioning + dust control; Crew: 1 operator, usable by non-specialists
WhichBot score: 8.3/10. Best for: MEP and general contractors on data centers, hospitals, hotels — any project with thousands of repetitive ceiling anchors.
Strengths: Most proven track record in the category (since 2020); Drills straight from BIM data — fewer errors, auto-documentation; Eliminates the most injury-prone task in MEP work
Weaknesses: Single-purpose: it drills and marks, nothing else; Needs clean BIM data and site prep to shine; ROI depends on repetitive, high-volume drilling scopes
Jaibot is the boring miracle of construction robotics: in the field since 2020 with contractors like Skanska and Bouygues, it is arguably the most proven robot on any job site. Its task is narrow and awful — drilling thousands of overhead anchor holes for mechanical, electrical, and plumbing installs, the single most injury-prone repetitive job in MEP work. Jaibot reads the BIM model, locates itself via total station, drills each hole to spec with built-in dust collection, color-marks it by trade, and syncs completion status back to the project system automatically. It runs a full shift on a charge and, per Hilti, doesn't need a specialist to operate.
Its honesty is its limitation: Jaibot drills and marks, and that's all. The value case lives on projects with high-volume repetitive scopes — data centers, hospitals, hotels, anywhere the same ceiling pattern repeats hundreds of times — where field results show weeks of ceiling work compressing into days. It also inherits construction robotics' universal dependency: clean, coordinated BIM data. Feed it a dirty model and you'll automate your errors at millimetre accuracy.
Our read: for MEP and general contractors with big repetitive drilling scopes, Jaibot is the closest thing to a no-regrets first robot — proven for half a decade, distributed through Hilti's familiar fleet channels, and attacking a task nobody's shoulders will miss.

## Dusty Robotics FieldPrinter (Construction & building robots)
URL: https://whichbot.net/robots/dusty-fieldprinter
Price: ~$3–5K/mo*. Dusty offers subscription pricing; industry estimates put it around $3,000–$5,000/month depending on configuration.
Availability: Subscription — Commercially available on subscription and widely used by US general contractors; layout errors cause roughly 30% of construction rework, which is the problem this robot attacks.
Specs: Task: Prints BIM layout onto slabs; Speed: ~75% faster than manual layout; Platform: Compact autonomous mobile printer; Sensing: Total-station positioning; Crew: 1 operator per unit
WhichBot score: 8.1/10. Best for: General contractors and framers on commercial interiors who want near-zero layout error across complex floor plans.
Strengths: Attacks rework, the most expensive waste in construction; Cuts layout time by roughly 75%; Design changes reach the slab within hours via BIM
Weaknesses: Value depends on clean, coordinated BIM models; Subscription economics favor busy contractors over occasional use; Interior slab work only — not a facade or exterior tool
FieldPrinter attacks the most expensive waste stream in construction with the least dramatic robot in this index. Layout errors cause roughly 30% of rework, and rework is the industry's silent budget killer — so Dusty built a compact autonomous printer that drives the slab and prints the entire BIM model onto the concrete at millimetre accuracy: walls, doors, penetrations, MEP runs, every trade's lines in one pass. Contractors report layout time dropping by about 75%, and when the design changes, the new layout is on the slab within hours instead of re-snapped by hand over days.
The subscription model (industry estimates put it around $3,000–$5,000 per month) fits how the value works: a busy GC running commercial interiors amortizes it across continuous projects, while occasional use pencils worse. Like Jaibot, its output quality is a mirror of your model quality — coordinated BIM in, flawless floors out; conflicts in, conflicts printed. And it's an interiors tool: slabs, not facades.
Our read: for general contractors and framers on complex commercial interiors, FieldPrinter has the clearest six-to-twelve-month payback in construction robotics, because it attacks rework — the cost everyone pays and nobody budgets. It's the recommendation we make when a contractor asks where to start.

## PaintJet Bravo (Construction & building robots)
URL: https://whichbot.net/robots/paintjet-bravo
Price: Service / quote. PaintJet operates a service model for large-scale industrial and commercial painting; pricing is quoted per project.
Availability: Service contract — In commercial service for industrial and large commercial exteriors — warehouses, plants, and big-box facades — using equipment already standard on job sites.
Specs: Task: Large-scale exterior & industrial painting; Speed: Multiples of a human crew on big walls; Platform: Mounts to standard lifts on site; Sensing: Vision-guided consistent coverage; Crew: Small supervising crew, off the wall
WhichBot score: 7.6/10. Best for: Owners and contractors of warehouses, plants, and large commercial facades that need repainting at scale.
Strengths: Uses standard lift equipment already on most sites; Consistent film thickness humans can't match at scale; Removes workers from elevated painting — the safety win
Weaknesses: Built for big flat expanses, not detailed trim work; Service model — not a machine you buy and own; Weather-dependent like all exterior coating work
Bravo answers a question facility owners actually have: who repaints half a million square feet of warehouse wall without putting a crew on booms for a month? PaintJet's robot mounts on the standard lifts already sitting on most job sites and coats large exteriors with machine-consistent film thickness — a spec humans genuinely can't match across acres of wall, and one that translates directly into coating lifespan. Crews supervise from the ground instead of working at height, which insurers and safety officers notice immediately.
It's a specialist sold as a service: big flat expanses are its territory, detailed trim work is not, and weather windows constrain it like any exterior coating operation. Pricing is quoted per project, which makes evaluation refreshingly simple — put your next repaint out to bid both ways and compare numbers that include the safety and schedule math.
Our read: for owners and contractors of warehouses, plants, and big-box facades, Bravo turns an unpleasant, hazardous, weather-hostage project into a quotable service with better coating consistency. It's the narrowest tool in our building index and completely unashamed of it — which is exactly why it works.

============================================================
# COMPARISONS

## Optimus Gen 3 vs NEO Gamma
URL: https://whichbot.net/compare/optimus-vs-neo
Optimus Gen 3: $25,000*, Waitlist, score 8.1/10. NEO Gamma: $20,000*, Home pilot, score 8.4/10.

## Optimus Gen 3 vs G1
URL: https://whichbot.net/compare/optimus-vs-g1
Optimus Gen 3: $25,000*, Waitlist, score 8.1/10. G1: $16,000, Buy today, score 7.2/10.

## Optimus Gen 3 vs Figure 03
URL: https://whichbot.net/compare/optimus-vs-figure
Optimus Gen 3: $25,000*, Waitlist, score 8.1/10. Figure 03: Not for sale, Commercial only, score 8.7/10.

## Optimus Gen 3 vs Apollo
URL: https://whichbot.net/compare/optimus-vs-apollo
Optimus Gen 3: $25,000*, Waitlist, score 8.1/10. Apollo: Lease only, Commercial only, score 7.9/10.

## Optimus Gen 3 vs Digit
URL: https://whichbot.net/compare/optimus-vs-digit
Optimus Gen 3: $25,000*, Waitlist, score 8.1/10. Digit: RaaS pricing, Commercial only, score 7.5/10.

## NEO Gamma vs G1
URL: https://whichbot.net/compare/neo-vs-g1
NEO Gamma: $20,000*, Home pilot, score 8.4/10. G1: $16,000, Buy today, score 7.2/10.

## NEO Gamma vs Figure 03
URL: https://whichbot.net/compare/neo-vs-figure
NEO Gamma: $20,000*, Home pilot, score 8.4/10. Figure 03: Not for sale, Commercial only, score 8.7/10.

## NEO Gamma vs Apollo
URL: https://whichbot.net/compare/neo-vs-apollo
NEO Gamma: $20,000*, Home pilot, score 8.4/10. Apollo: Lease only, Commercial only, score 7.9/10.

## NEO Gamma vs Digit
URL: https://whichbot.net/compare/neo-vs-digit
NEO Gamma: $20,000*, Home pilot, score 8.4/10. Digit: RaaS pricing, Commercial only, score 7.5/10.

## G1 vs Figure 03
URL: https://whichbot.net/compare/g1-vs-figure
G1: $16,000, Buy today, score 7.2/10. Figure 03: Not for sale, Commercial only, score 8.7/10.

## G1 vs Apollo
URL: https://whichbot.net/compare/g1-vs-apollo
G1: $16,000, Buy today, score 7.2/10. Apollo: Lease only, Commercial only, score 7.9/10.

## G1 vs Digit
URL: https://whichbot.net/compare/g1-vs-digit
G1: $16,000, Buy today, score 7.2/10. Digit: RaaS pricing, Commercial only, score 7.5/10.

## Figure 03 vs Apollo
URL: https://whichbot.net/compare/figure-vs-apollo
Figure 03: Not for sale, Commercial only, score 8.7/10. Apollo: Lease only, Commercial only, score 7.9/10.

## Figure 03 vs Digit
URL: https://whichbot.net/compare/figure-vs-digit
Figure 03: Not for sale, Commercial only, score 8.7/10. Digit: RaaS pricing, Commercial only, score 7.5/10.

## Apollo vs Digit
URL: https://whichbot.net/compare/apollo-vs-digit
Apollo: Lease only, Commercial only, score 7.9/10. Digit: RaaS pricing, Commercial only, score 7.5/10.

## Ozmo vs Spot
URL: https://whichbot.net/compare/ozmo-vs-spot
Ozmo: RaaS / quote, Service contract, score 8.6/10. Spot: From $74,500, Buy today, score 8.2/10.

## Ozmo vs Hadrian X
URL: https://whichbot.net/compare/ozmo-vs-hadrianx
Ozmo: RaaS / quote, Service contract, score 8.6/10. Hadrian X: Walls-as-a-Service, Service contract, score 8/10.

## Ozmo vs Jaibot
URL: https://whichbot.net/compare/ozmo-vs-jaibot
Ozmo: RaaS / quote, Service contract, score 8.6/10. Jaibot: Quote / lease, Lease / fleet, score 8.3/10.

## Ozmo vs FieldPrinter
URL: https://whichbot.net/compare/ozmo-vs-dusty
Ozmo: RaaS / quote, Service contract, score 8.6/10. FieldPrinter: ~$3–5K/mo*, Subscription, score 8.1/10.

## Ozmo vs Bravo
URL: https://whichbot.net/compare/ozmo-vs-paintjet
Ozmo: RaaS / quote, Service contract, score 8.6/10. Bravo: Service / quote, Service contract, score 7.6/10.

## Spot vs Hadrian X
URL: https://whichbot.net/compare/spot-vs-hadrianx
Spot: From $74,500, Buy today, score 8.2/10. Hadrian X: Walls-as-a-Service, Service contract, score 8/10.

## Spot vs Jaibot
URL: https://whichbot.net/compare/spot-vs-jaibot
Spot: From $74,500, Buy today, score 8.2/10. Jaibot: Quote / lease, Lease / fleet, score 8.3/10.

## Spot vs FieldPrinter
URL: https://whichbot.net/compare/spot-vs-dusty
Spot: From $74,500, Buy today, score 8.2/10. FieldPrinter: ~$3–5K/mo*, Subscription, score 8.1/10.

## Spot vs Bravo
URL: https://whichbot.net/compare/spot-vs-paintjet
Spot: From $74,500, Buy today, score 8.2/10. Bravo: Service / quote, Service contract, score 7.6/10.

## Hadrian X vs Jaibot
URL: https://whichbot.net/compare/hadrianx-vs-jaibot
Hadrian X: Walls-as-a-Service, Service contract, score 8/10. Jaibot: Quote / lease, Lease / fleet, score 8.3/10.

## Hadrian X vs FieldPrinter
URL: https://whichbot.net/compare/hadrianx-vs-dusty
Hadrian X: Walls-as-a-Service, Service contract, score 8/10. FieldPrinter: ~$3–5K/mo*, Subscription, score 8.1/10.

## Hadrian X vs Bravo
URL: https://whichbot.net/compare/hadrianx-vs-paintjet
Hadrian X: Walls-as-a-Service, Service contract, score 8/10. Bravo: Service / quote, Service contract, score 7.6/10.

## Jaibot vs FieldPrinter
URL: https://whichbot.net/compare/jaibot-vs-dusty
Jaibot: Quote / lease, Lease / fleet, score 8.3/10. FieldPrinter: ~$3–5K/mo*, Subscription, score 8.1/10.

## Jaibot vs Bravo
URL: https://whichbot.net/compare/jaibot-vs-paintjet
Jaibot: Quote / lease, Lease / fleet, score 8.3/10. Bravo: Service / quote, Service contract, score 7.6/10.

## FieldPrinter vs Bravo
URL: https://whichbot.net/compare/dusty-vs-paintjet
FieldPrinter: ~$3–5K/mo*, Subscription, score 8.1/10. Bravo: Service / quote, Service contract, score 7.6/10.

============================================================
# BUYER'S GUIDES

## The best home robots of 2026, ranked.
URL: https://whichbot.net/guides/best-home-robots
Target topics: best home robot 2026, best humanoid robot, best robot for home use
A "best robot" list is only useful if it starts from one brutal filter: can you actually get it? Plenty of humanoids demo beautifully on stage and then ship exclusively to warehouses. This ranking weighs real-world availability, safety around people, genuine capability, and value — in that order.
Short version: if you want a robot in your home this year, one pick stands clearly apart. If you can wait, the calculus changes.
### How we ranked them
Availability is weighted heaviest because a robot you cannot buy scores zero in your living room, no matter how good its demo reel is. Safety comes second — a home robot shares space with children, pets, and grandparents, which is a completely different engineering problem from a fenced warehouse cell. Capability and value round out the score.
Commercial-only machines like Figure 03, Apollo, and Digit are excluded from the ranking despite impressive specs — they're covered in our full reviews and comparisons for readers tracking where home capability comes from next.
### What a home robot can actually do in 2026
Set expectations honestly: today's best home humanoids handle light tidying, carrying, fetching, opening doors, and monitoring — with more complex chores like laundry folding demonstrated but slow, and some tasks still assisted by remote teleoperation. The trajectory is steep, but the marketing videos run ahead of the Tuesday-afternoon reality.
### FAQ
Q: What is the best home robot in 2026?
A: For most households, the 1X NEO Gamma is the best home robot in 2026 — it is the only full-size humanoid actually deployed in consumer homes, with a soft-body design built for safety around people and pets. Tesla Optimus is the one to wait for; Unitree G1 is the best you can order today.
Q: How much does a good home robot cost?
A: Realistic 2026 pricing runs $16,000 (Unitree G1, orderable now) to an estimated $20,000–$30,000 for home-oriented humanoids like 1X NEO and Tesla Optimus. Prices are expected to fall as production scales.
Q: Can home robots do chores?
A: Partially. Current home humanoids handle light tidying, carrying, fetching and monitoring. Complex chores like folding laundry are demonstrated but slow, and some tasks still rely on remote human assistance. Capability is improving quarter over quarter.

## Humanoid robots for sale: what you can actually buy.
URL: https://whichbot.net/guides/humanoid-robots-for-sale
Target topics: humanoid robots for sale, buy humanoid robot, humanoid robot for sale under $20,000
Search "humanoid robots for sale" and you'll drown in stage demos, renders, and robots that are exclusively leased to Fortune 500 warehouses. This page tracks the only thing that matters to a buyer: which humanoids you can genuinely acquire, at what price, through what process — updated as availability changes.
### The ones you cannot buy (and why that matters)
Figure 03, Apptronik Apollo, and Agility Digit are arguably the three most capable humanoids in the world — and none is sold to consumers at any price. They deploy through enterprise contracts and robots-as-a-service leases. We track them anyway, because today's warehouse capability is a preview of what reaches your kitchen in two to three years.
### Import and gray-market warnings
A growing number of resellers list Chinese humanoids at tempting prices. Before wiring money: confirm the seller is an authorized distributor, factor in import duties and shipping (often thousands of dollars), and understand that warranty service may require shipping the unit overseas. A $2,000 discount is not worth an unserviceable robot.
### FAQ
Q: Can I buy a humanoid robot right now?
A: Yes. The Unitree G1 is orderable today at a $16,000 list price. The 1X NEO is accessible through a home pilot program. Most other humanoids — including Figure 03 and Apptronik Apollo — are commercial-only and not sold to consumers.
Q: Is there a humanoid robot for sale under $20,000?
A: Yes — the Unitree G1 at $16,000 is the most prominent humanoid under $20,000 that actually ships to buyers. Smaller research and hobbyist platforms exist below that price but are not household robots.
Q: When will Tesla Optimus be for sale?
A: Tesla has not announced a consumer on-sale date. Production lines are being converted and early units serve Tesla's own operations. Public statements target eventual pricing around $20,000–$30,000 at volume.

## The cheapest humanoid robots that actually exist.
URL: https://whichbot.net/guides/cheapest-humanoid-robot
Target topics: cheapest humanoid robot, affordable humanoid robot
The floor for a real, walking, full-body humanoid you can order today is around $16,000 — the Unitree G1. Below that price, listings are usually miniature hobby kits, research torsos, or vaporware. Here's what the budget end of the market genuinely looks like, and where prices go next.
### Why prices are about to fall
Three forces are compressing humanoid prices: Chinese manufacturers scaling production aggressively (Unitree's IPO signals exactly this), Tesla applying automotive cost engineering to Optimus, and component costs — actuators, sensors, compute — dropping as volumes rise. The pattern to expect resembles early EVs: premium first, then a steep glide toward mass-market pricing over several years.
Practical advice: if you don't have a burning use case today, waiting 12–18 months will likely buy you significantly more robot per dollar. If tinkering IS the use case, the G1 is already good value.
### The subscription wildcard
Several makers have floated robot-as-a-subscription models for homes — a monthly fee instead of a five-figure purchase. If that lands, "cheapest" stops meaning sticker price and starts meaning monthly cost, the same shift that happened with phones. We'll track subscription offers on this page as they launch.
### FAQ
Q: What is the cheapest humanoid robot in 2026?
A: The Unitree G1 at a $16,000 list price is the cheapest full-size humanoid robot consumers can actually order in 2026. Cheaper listings are typically miniature kits, research platforms, or not real products.
Q: Will humanoid robots get cheaper?
A: Almost certainly. Manufacturing scale-ups in China, Tesla's automotive-style cost engineering, and falling component prices are all pushing the same direction. Analysts widely expect home humanoid prices to decline meaningfully over the next several years.
Q: Are cheap humanoid robots worth buying?
A: For tinkerers and developers, yes — the G1 is real hardware with a strong community. For someone who wants a household helper with no technical effort, budget options will frustrate you; wait for home-oriented models to mature.

## What humanoid robots cost in 2026.
URL: https://whichbot.net/guides/humanoid-robot-price
Target topics: humanoid robot price, home robot price, humanoid robot cost, how much will home robots cost
Humanoid pricing in 2026 spans from $16,000 to "call our enterprise sales team." This page keeps every known price and credible estimate in one place — clearly labeled, dated, and updated when the market moves.
### What you're actually paying for
Roughly speaking, a humanoid's cost stacks up from actuators (the single biggest line item — a full-size humanoid carries dozens of them), sensors and cameras, onboard compute for AI, batteries, and structure. That's why payload and dexterity correlate so tightly with price: more capable joints mean more expensive actuators, everywhere.
The costs that don't show on the sticker: electricity is trivial (comparable to a gaming PC), but consider insurance, potential subscription software fees, and repairs — there is no corner robot mechanic yet, and out-of-warranty service on a new category is an unknown.
### Where prices go from here
The credible bull case says home humanoids follow flat-screen TVs: luxury pricing for early adopters, then a relentless slide as manufacturing scales. The bear case says complex electromechanical products (cars, for instance) plateau rather than collapse in price. Our read: expect the entry point to drop toward the low five figures within a few years, with capable home models holding a premium above that.
### FAQ
Q: How much does a humanoid robot cost in 2026?
A: Consumer-relevant humanoid robots range from $16,000 (Unitree G1, orderable today) to an estimated $20,000–$30,000 for home-oriented models like 1X NEO and Tesla Optimus. The most capable commercial humanoids are not sold at any consumer price.
Q: How much will home robots cost in the future?
A: Most credible projections see entry prices falling toward the low five figures as production scales, following the pattern of other new hardware categories. Subscription models may also replace upfront purchase for home robots.
Q: Why are humanoid robots so expensive?
A: Actuators are the main cost driver — a full-size humanoid contains dozens of precision motorized joints — followed by sensors, AI compute, and batteries. Prices fall as these components scale in volume.

## The robot butler is real now. Sort of.
URL: https://whichbot.net/guides/robot-butler
Target topics: robot butler price, robot maid, personal robot assistant, home robot assistant
The phrase people actually search — "robot butler," "robot maid" — is the honest one. Nobody dreams of a "general-purpose bipedal manipulation platform"; they want something that tidies up, fetches things, and maybe judges their guests silently. Here's how close 2026 gets you.
### What a 2026 robot butler can and cannot do
Can: navigate your home, pick up and carry objects, tidy loose items, fetch and deliver, open doors, respond to voice, keep an eye on things while you're out. Cannot (or only slowly, or with remote assistance): cook a full meal, fold laundry at human speed, clean bathrooms, handle fragile chaos like a toddler's aftermath at pace. The gap closes every quarter, but buy for what it does today, not the demo reel.
### Butler economics
A useful frame: a $20,000 robot amortized over five years is roughly $11 a day before repairs. Against the cost of human help for even a few hours a week, the math starts getting interesting well before the robot matches human capability — which is precisely why the industry is racing at this category.
### FAQ
Q: How much does a robot butler cost?
A: The closest real products cost $16,000–$30,000 in 2026. The 1X NEO (estimated ~$20,000) is the nearest thing to a purchasable robot butler, currently reaching homes via a pilot program.
Q: Is there a real robot maid?
A: Early versions, yes. The 1X NEO operates in real homes doing light tidying, carrying, and fetching. Full maid-level cleaning — bathrooms, dishes, fast laundry — remains beyond current home robots.
Q: What can a personal robot assistant do in 2026?
A: Navigate a home, carry and fetch objects, tidy loose items, open doors, respond to voice commands, and monitor the house. Complex chores are demonstrated but slow, and some tasks still use remote human assistance.

## Robots for elderly care: the honest 2026 guide.
URL: https://whichbot.net/guides/robots-for-elderly-care
Target topics: robot companion for elderly, robot for elderly care at home
This is the most consequential use case in home robotics, and the one where honest expectations matter most. Families aren't buying a gadget — they're weighing whether technology can help a parent stay independent at home. Here's what 2026 robots genuinely offer for that, and where they fall short.
### What robots genuinely help with
The realistic 2026 value: fetching and carrying (reducing fall-risk trips), reminders and routines, monitoring with alerts to family, companionship through conversation, and being a set of hands for small tasks. These are meaningful — falls while carrying things and missed medications are two of the biggest independence-enders.
What no home robot does in 2026: physical transfer assistance (helping someone out of bed or a bath), medical care, or emergency physical intervention. Any marketing implying otherwise deserves deep skepticism.
### Safety and dignity considerations
Involve the person in the decision — a robot imposed on a resistant parent becomes an expensive coat rack. Weight matters: a soft, light robot poses far less risk in a collision or fall than an industrial-grade machine. And think through the camera question openly: monitoring that feels like care to an adult child can feel like surveillance to a parent. Set the rules together.
### FAQ
Q: Can a robot take care of an elderly person at home?
A: Not fully. 2026 home robots assist — fetching, carrying, reminders, monitoring, companionship — but cannot provide physical care like transfer assistance or emergency intervention. They complement human care rather than replace it.
Q: What is the best robot for elderly care in 2026?
A: The 1X NEO is currently best suited: its soft-body, lightweight, safety-first design is appropriate for shared space with older adults, and it is genuinely deployed in homes. Treat it as an assistant, not a caregiver.
Q: How much does an elderly care robot cost?
A: Home humanoids suited to assisting older adults run an estimated $20,000–$30,000 in 2026, with subscription models possibly arriving. Compare against the monthly cost of in-home help when evaluating value.

## What humanoid robots can actually do.
URL: https://whichbot.net/guides/what-can-humanoid-robots-do
Target topics: what can humanoid robots do, robot that does household chores, can robots cook
Every capability claim in this guide is sorted into three honest buckets: things robots do reliably today, things they do slowly or with remote human assistance, and things that remain demo-only. If a video amazed you recently, this page tells you which bucket it belongs in.
### Reliable today
Walking and navigating homes and workplaces, including stairs and door handles. Picking up, carrying, and delivering objects up to their payload rating. Sorting and moving items — the backbone of warehouse deployments processing hundreds of thousands of packages. Voice interaction, monitoring, and patrolling. In factories, repetitive manipulation tasks now run genuinely unattended.
### Real but slow (or assisted)
Laundry folding, loading dishwashers, wiping surfaces, and tidying cluttered rooms all exist on real robots — at a fraction of human speed, and sometimes with a remote human operator assisting the hard parts. This "teleoperation" layer is the industry's open secret: it's how robots handle edge cases while the AI learns. It's also how they improve, so this bucket empties into the reliable bucket over time.
### Still demo-only
Cooking a full meal (chopping, stovetop judgment, plating) remains a staged showcase, not a shipping feature. So do bathroom cleaning, childcare of any kind, and complex repairs. When you see these in a video, assume careful choreography, favorable takes, or teleoperation unless proven otherwise.
### FAQ
Q: Can humanoid robots do household chores?
A: Partially. Carrying, fetching, tidying loose items, and monitoring are reliable. Laundry folding and dish loading work but slowly, sometimes with remote human assistance. Bathroom cleaning and cooking remain effectively demo-only in 2026.
Q: Can robots cook in 2026?
A: Not really. Full meal preparation — chopping, stovetop control, plating — remains a staged demonstration rather than a shipping capability on any home humanoid in 2026.
Q: Is there a robot that does laundry?
A: Robots can fold laundry in 2026, but slowly and imperfectly. It is a genuine capability under active development, not yet a practical replacement for doing it yourself. See our dedicated laundry robot guide for details.

## The laundry robot: how close are we?
URL: https://whichbot.net/guides/robot-that-does-laundry
Target topics: robot that does laundry
Laundry folding has become the unofficial benchmark of home robotics — soft, crumpled, unpredictable fabric is one of the hardest manipulation problems in the field, which is exactly why every robot company shows off folding videos. Here's the truth behind them.
### The state of robot laundry in 2026
Multiple humanoids — including 1X's NEO and Figure's robots — have demonstrated real laundry handling: picking items from a basket, folding shirts, moving loads between machines. The catch is speed and reliability. A human folds a shirt in seconds; robots take substantially longer and fumble edge cases (fitted sheets defeat robots as thoroughly as they defeat humans). Some demos also involve remote human assistance for the hardest grasps.
That said, this is the fastest-improving capability in home robotics, because it's the one every company is being measured on. The realistic near-term product isn't "laundry done invisibly" — it's "the robot slowly works through the basket while you're at work, and you don't care that it took two hours."
### Why laundry is so hard for robots
Rigid objects have predictable shapes; a dropped towel has effectively infinite configurations. Solving deformable-object manipulation requires exactly the vision-language-action AI models the industry is racing to build — which is why laundry progress is a leading indicator for every other soft-hands household task, from making beds to packing bags.
### FAQ
Q: Is there a robot that does laundry in 2026?
A: Partially. Humanoids like the 1X NEO have demonstrated genuine laundry folding and handling, but at much slower than human speed and with occasional remote human assistance. It is an emerging capability, not yet a solved chore.
Q: When will robots be able to do laundry properly?
A: Laundry manipulation is the fastest-improving home robot capability. Most industry roadmaps suggest practical, unattended laundry handling within the next few years as vision-language-action AI models mature — though "slow but unattended" arrives before "fast."

## Are home robots safe around your family?
URL: https://whichbot.net/guides/are-home-robots-safe
Target topics: are home robots safe
Three different questions hide inside "is it safe": Can it physically hurt someone? What does it see and where does that data go? Can someone hack it? Each deserves a straight answer, because a humanoid in your home is a new kind of houseguest — heavy, camera-covered, and internet-connected.
### Physical safety
The design philosophy split matters enormously here. Home-first robots like 1X's NEO are deliberately lightweight (~30 kg) with soft, compliant bodies — engineered so that a collision or fall is a bump, not an injury. Industrial-lineage humanoids run 55–75 kg of rigid metal, which is why they live behind warehouse safety protocols, not in living rooms. Rule of thumb: for a home with kids or pets, weight and compliance matter more than any capability spec.
Every credible home robot also layers software limits — speed caps around people, force limits on grips, stop-on-contact behavior. These work, but they're software; the physics of a light soft robot is the deeper safeguard.
### Privacy and security
A home humanoid is a walking sensor array: cameras, microphones, depth sensors, mapping your house continuously. Before buying, get answers to: What footage leaves the device? Is remote teleoperation possible, and under what consent? Can you review and delete recordings? Where is data stored and for how long? Reputable makers publish this; evasiveness is itself an answer.
On hacking: treat a robot like any high-stakes connected device. Strong unique password, isolated network if possible, automatic security updates, and a maker with a real security team. The industry knows a headline breach would set the category back years — which, cynically but usefully, means the serious players invest heavily here.
### FAQ
Q: Are home robots safe around children?
A: Home-first designs like the 1X NEO — lightweight, soft-bodied, speed-limited around people — are engineered specifically for this and are operating in real family homes. Heavier industrial-style humanoids are not appropriate for homes with children.
Q: Do home robots record you?
A: Home humanoids continuously use cameras and microphones to function. What gets stored or transmitted varies by maker — review each company's data policy for footage retention, remote-access consent, and deletion rights before buying.
Q: Can home robots be hacked?
A: Any connected device can be attacked. Mitigate with strong unique passwords, network isolation, and automatic updates — and prefer makers with published security practices. No major home humanoid breach has been reported as of mid-2026.

## When can you actually get a home robot?
URL: https://whichbot.net/guides/when-will-home-robots-be-available
Target topics: when will home robots be available, when can I buy tesla optimus
The answer is "now, sort of, depending" — which deserves unpacking. Home robot availability in 2026 is a staircase: one step you can stand on today, several steps with dates attached, and a landing (true mass market) whose timing is the biggest open question in consumer tech.
### Available now (mid-2026)
The Unitree G1 is orderable today — a real humanoid, though a developer platform more than a household helper. The 1X NEO is in expanding home pilot programs; joining the pilot is currently the realistic path to having a home-oriented humanoid actually living in your house this year.
### The next 12–24 months
Expect: NEO's pilot widening toward open availability, Chinese manufacturers (flush with IPO capital) pushing consumer-priced models westward, and — the wildcard — Tesla potentially opening Optimus reservations. Tesla's production lines are being converted now, but history counsels patience: Tesla timelines are directional, not contractual. Robotics timelines in general slip; plan around capabilities you can verify, not launch promises.
Our honest read: 2027 is when "just buy one online" plausibly becomes true for a home-oriented humanoid in the US, with 2026 belonging to pilots, waitlists, and developer units.
### FAQ
Q: When will home robots be available to buy?
A: Limited availability exists now: Unitree G1 is orderable today and 1X NEO reaches homes via pilot programs. Broad "add to cart" availability for home-oriented humanoids most plausibly arrives around 2027 as pilots open up and production scales.
Q: When can I buy a Tesla Optimus?
A: No consumer date is announced. Tesla is converting production lines and deploying early units internally first. Based on Tesla's vehicle playbook, a reservation system would likely precede deliveries — watch for that signal.
Q: Should I wait to buy a home robot?
A: If you have no urgent use case, waiting 12–18 months likely buys more capability per dollar as production scales. If early access itself is the point — tinkering, content, or genuine assistance needs — the G1 and NEO pilot are real options today.

## How humanoid robots work, in plain English.
URL: https://whichbot.net/guides/how-do-humanoid-robots-work
Target topics: how do humanoid robots work
You don't need an engineering degree to be a smart robot buyer — but understanding three building blocks (muscles, senses, brain) explains almost every price difference and capability gap on the market. Here's the five-minute version.
### Muscles: actuators
Every joint in a humanoid is an actuator — a motor-gearbox package that converts electricity into precise motion. A full-size humanoid carries dozens, and they dominate the cost of the machine. This is also where design philosophy shows: rigid, powerful actuators give industrial robots their 20–25 kg payloads; softer, compliant actuation gives home robots like NEO their safety around people. When you compare payload specs on our site, you're really comparing actuator budgets.
### Senses: cameras and everything else
Humanoids see mostly through cameras, augmented by depth sensors, microphones, joint-position encoders, and force sensors in hands and feet. The industry has largely converged on the "cameras plus AI" approach that Tesla champions — put simply, the same bet as self-driving: rich vision plus a smart brain beats exotic sensors.
### Brain: foundation models
The revolution that made 2026's robots possible isn't mechanical — it's AI. Modern humanoids run "vision-language-action" models: the same family of technology behind chatbots, trained instead to translate what the robot sees and hears into motor commands. This is why capabilities improve via software update, why robots learn from watching humans, and why one company's warehouse experience transfers to your kitchen. It's also why remote human "teleoperation" exists — humans handle the edge cases, and the AI trains on those recordings until it doesn't need to.
### FAQ
Q: How do humanoid robots work?
A: Three systems: actuators (motorized joints acting as muscles), sensors (primarily cameras, plus depth, force and audio), and an AI brain — vision-language-action models that translate perception into movement. Capabilities improve largely through software updates.
Q: Why do humanoid robots need AI?
A: Pre-programmed motion only works in fixed environments. Homes are unpredictable, so humanoids use AI foundation models to interpret what they see and decide movements in real time — the same technological family as modern chatbots, adapted for physical action.
Q: What is teleoperation in robotics?
A: Remote human operators assisting a robot with tasks its AI cannot yet handle. The industry uses it both to serve customers today and to generate training data — each assisted task teaches the AI to handle it autonomously next time.

## Is a humanoid robot worth it yet?
URL: https://whichbot.net/guides/are-humanoid-robots-worth-it
Target topics: are humanoid robots worth it
The honest answer is a matrix, not a yes. For three kinds of buyers, 2026 humanoids are already worth it; for most households, waiting is mathematically the better move. Here's the breakdown we'd give a friend.
### Worth it now, for three buyers
Tinkerers and developers: a $16,000 G1 is a frontier-technology platform with a thriving community — for this buyer, the robot is the hobby, and it delivers. Early-access households: if having assistance early matters (mobility limitations, or simply valuing the head start), the NEO pilot delivers genuine daily value today. Content creators and businesses: a humanoid pays for itself in attention or labor in ways a household budget can't match.
### The waiting math for everyone else
Two curves are moving against buying today: prices are heading down as manufacturing scales, and capabilities are heading up via software. A robot bought in 12–18 months will likely cost less and do more — the classic early-adopter tax. Against that, weigh what a year of assistance is worth to you now. For most households where the robot would be a fascinating helper rather than a needed one, the tax isn't worth paying yet.
Our practical suggestion: take the quiz to see which robot fits you, join its waitlist or mailing list (free), and let the market ripen while you hold a place in line.
### FAQ
Q: Are humanoid robots worth buying in 2026?
A: For tinkerers, early-access households, and businesses — yes, real value exists today. For most households, prices falling and capabilities rising make waiting 12–18 months the mathematically better choice.
Q: Do home robots save money?
A: Framed against paid human help, a $20,000 robot over five years costs roughly $11/day — competitive with even a few hours of weekly help. But 2026 robots replace only a slice of what human help does, so the savings case is early.
Q: What is the early-adopter tax on robots?
A: Buying now means paying more for less capability than buyers a year later — because prices fall with manufacturing scale while capabilities improve through software updates. Whether early access justifies that premium is the core "worth it" question.

## Every humanoid robot company that matters.
URL: https://whichbot.net/guides/humanoid-robot-companies
Target topics: humanoid robot companies
The humanoid industry sorts into three camps: the consumer-bound (building for your home), the industrial-first (building for warehouses, watching your home), and the platform players (selling hardware for others to build on). Knowing which camp a company is in tells you whether its announcements matter to you as a buyer.
### The pattern to watch
Capability flows downhill from warehouses to homes: skills proven in commercial deployment (Figure's package handling, Apollo's shift work) reappear two to three years later in consumer products. That's why we track commercial-only companies on a consumer site — today's warehouse robot is the preview of your 2028 household model. Meanwhile Chinese manufacturers, led by Unitree's post-IPO expansion, exert constant downward pressure on prices across the whole market. The consumer wins either way.
### FAQ
Q: What are the biggest humanoid robot companies?
A: The most consequential in 2026: Tesla (Optimus), Figure AI, 1X Technologies (NEO), Unitree (first humanoid IPO), Apptronik (Apollo), and Agility Robotics (Digit) — spanning consumer, industrial, and platform strategies.
Q: Which robot companies sell to consumers?
A: Unitree sells the G1 directly today; 1X reaches consumers through its NEO home pilot; Tesla intends consumer sales but hasn't opened them. Figure, Apptronik, and Agility are commercial-only.
Q: Who is the leader in humanoid robots?
A: Depends on the axis: Figure leads on demonstrated capability, 1X on actual home deployment, Unitree on price and volume, Tesla on manufacturing potential. No single company leads on all fronts in 2026.

## The new robots of 2026, tracked.
URL: https://whichbot.net/guides/new-humanoid-robots-2026
Target topics: new humanoid robots 2026, humanoid robot news
This is the running log of 2026's humanoid milestones — the launches, price moves, and production signals that actually change what you can buy. Bookmark it; it updates as the year unfolds.
### 2026 so far: the milestones that matter
Unitree completed the industry's first humanoid IPO — significant less for the ticker than for what the capital does: China's volume leader now has public-market fuel for a price war. Tesla converted Model S/X production lines to Optimus manufacturing, the strongest physical evidence yet that its robot ambitions are industrial reality rather than keynote material. 1X's NEO home pilot kept expanding, quietly normalizing the strangest idea in consumer tech: a humanoid that just... lives at your house. And Figure's logistics deployments crossed the hundreds-of-thousands-of-packages threshold with a clean reliability record — the capability high-water mark the home market inherits next.
### What to watch next
The three signals that would change our buying advice overnight: Tesla opening any form of Optimus reservation system; 1X converting its pilot into open ordering; and any credible sub-$10,000 humanoid from a Chinese manufacturer reaching Western consumers with real support. Any of those flips this from an early-adopter market to a mainstream one — and this page will say so the day it happens.
### FAQ
Q: What new humanoid robots launched in 2026?
A: The headline 2026 developments: Unitree's industry-first IPO, Tesla converting vehicle production lines to build Optimus, 1X expanding its NEO home pilot, and Figure scaling commercial deployments past hundreds of thousands of packages handled.
Q: What is the newest home robot?
A: The 1X NEO Gamma is the newest robot genuinely operating in consumer homes as of mid-2026, via 1X's expanding pilot program.

## Robots are cleaning skyscrapers right now.
URL: https://whichbot.net/guides/window-cleaning-robots-for-buildings
Target topics: window cleaning robot skyscraper, facade cleaning robot, robot window washer high rise, automated window cleaning system
This is no longer a concept video. A robotic system is cleaning the windows of a 45-story Class A tower in Manhattan full-time, riding the building’s existing washing cradle and working roughly three times faster than a human crew. For building owners and facade contractors, the question has shifted from "is this real" to "does it pencil for my building."
### How the economics work
Facade cleaning robots are sold as robot-as-a-service — you contract the cleaning outcome, not the machine. Pricing scales with facade area and schedule. The value stack: roughly 3× the cleaning speed, one operator instead of a suspended crew, no fall exposure on the wall (which insurers notice), and consistent quality on every pass. The labor context matters too: three-quarters of US window cleaners are over 40, and the trade is not replenishing — automation here is filling a shortage more than displacing workers.
Best-fit buildings today: towers that already have a building maintenance unit (BMU/cradle), glass-dominant facades, and recurring cleaning contracts. Complex ornamental facades and buildings without roof rigs remain human territory for now.
### What building owners should ask vendors
Four questions cut through the pitch: Does it work with our existing BMU or require rigging changes? What are the wind and weather operating limits? Who certifies and supplies the rooftop operator, and what does local regulation require? And what does the service contract cost against our current cleaning contract on a per-year basis — including the mobilization visits? Get those four answered in writing and you can make a real decision.
### FAQ
Q: Are there robots that clean skyscraper windows?
A: Yes — in commercial service today. Skyline Robotics’ Ozmo cleans a 45-story Manhattan tower full-time using a robotic arm on the building’s existing washing cradle, working about three times faster than traditional crews, with expansion underway in London.
Q: How much does a window cleaning robot cost?
A: Facade cleaning robots like Ozmo are sold as a service, not hardware — pricing is quoted per building based on facade size and cleaning schedule. Owners should compare the annual service quote against their current human-crew cleaning contract.
Q: Do facade robots replace window washers?
A: Partially. One certified rooftop operator supervises the robot instead of a multi-person suspended crew. The industry frames it as addressing a real labor shortage — most US window cleaners are over 40 and few young workers are entering the trade.

## Construction robots that actually work.
URL: https://whichbot.net/guides/construction-robots
Target topics: construction robots, construction robots 2026, robots in construction
Construction robotics crossed a line recently: from demo projects to "things contractors bring back on the right jobs." With roughly half a million unfilled construction positions in the US and the average skilled mason now over 55, robots are arriving as labor relief, not experiments. Here’s what’s genuinely proven on site in 2026.
### Where the ROI is real (and where it is not)
Industry payback patterns are consistent: inspection and scanning robots pay back in under six months; layout robots in six to twelve months via rework reduction; bricklaying systems run three-to-five-year paybacks but compress schedules dramatically. The hidden ROI everyone underestimates is schedule compression — finishing a $50M project two weeks early saves six figures in carry costs alone.
The consistent failure mode: deploying robots on a disorganized site. Every successful construction robot is BIM-driven — it reads the digital model directly. If your models are dirty or your workflows are paper-based, fix that before renting a robot, or you’ll automate your errors at millimetre accuracy.
### How to start without betting the company
The proven entry path: pick the trade shortage causing your worst schedule pain and pilot the robot that attacks it — masonry pain points to Hadrian X, MEP drilling to Jaibot, layout bottlenecks to FieldPrinter or HP SitePrint, documentation gaps to Spot. Nearly every vendor offers pilot programs, and most price as subscription or service, so the downside of a pilot is one project’s learning curve, not a capital write-off.
### FAQ
Q: What robots are used in construction in 2026?
A: The proven categories: layout printing robots (Dusty FieldPrinter, HP SitePrint), overhead drilling robots (Hilti Jaibot), bricklaying systems (FBR Hadrian X), site inspection robots (Boston Dynamics Spot), plus autonomous heavy equipment retrofits and drywall-finishing robots.
Q: Do construction robots replace workers?
A: Mostly they fill gaps: the US has roughly 500,000 unfilled construction positions and an aging skilled workforce. Most systems require human operators or supervisors, shifting workers into higher-skilled roles rather than eliminating crews.
Q: What is the ROI on construction robots?
A: It varies by category: inspection robots often pay back in under 6 months, layout robots in 6–12 months through rework reduction, and bricklaying systems over 3–5 years — with schedule compression as the biggest hidden return.

## Every construction robotics company worth knowing.
URL: https://whichbot.net/guides/construction-robot-companies
Target topics: construction robotics companies, building robot companies
Construction robotics sorts by trade, not by hype: each serious company owns one painful, repetitive job. Knowing who owns which trade tells you exactly who to call when a specific shortage is wrecking your schedule.
### Names beyond our index
Worth tracking even though we don’t index them yet: Canvas (drywall finishing, acquired by JLG in early 2026 — a signal that equipment giants are buying in), Built Robotics (autonomous retrofits for excavators, roughly $150K–300K per machine), HP SitePrint (layout, competing with Dusty), Construction Robotics’ SAM100 (semi-automated masonry around $500K), Brokk (remote demolition), TyBOT (rebar tying), and Monumental (bricklaying, Netherlands). The JLG-Canvas deal and Hilti’s decade of investment point the same direction: construction robotics is consolidating into the industry’s existing equipment channels — which means easier procurement for contractors every year.
### FAQ
Q: Who are the leading construction robotics companies?
A: By trade: Skyline Robotics (facade cleaning), Hilti (Jaibot drilling), FBR (Hadrian X bricklaying), Dusty Robotics (layout), Boston Dynamics (Spot inspection), PaintJet (coatings), plus Canvas (drywall, now JLG-owned), Built Robotics (autonomous equipment), and HP (SitePrint).
Q: How do construction robot companies sell — purchase or service?
A: Mostly service and subscription: Ozmo and Hadrian X are robot/walls-as-a-service, Dusty is subscription, Jaibot goes through Hilti fleet programs. Spot is a straight purchase from $74,500. The service model keeps contractor risk low.

## The robot bricklayers are hired.
URL: https://whichbot.net/guides/bricklaying-robots
Target topics: bricklaying robot, robot bricklayer, Hadrian X
Masonry is the trade where the labor math is most brutal — the average skilled mason is over 55, and there are fewer licensed masons in the US than at any point since 1950. That’s why bricklaying became robotics’ marquee construction challenge, and 2026 is the year the flagship machine went genuinely commercial.
### How robot bricklaying actually works
Hadrian X doesn’t mimic a human mason. It uses large-format blocks (roughly twice standard brick size), places them with a laser-stabilized boom that counteracts wind and vibration in real time, and bonds them with a rapid construction adhesive rather than mortar — which FBR claims outperforms traditional mortar. The machine cuts blocks to size on board and builds directly from the CAD model, leaving openings for windows and doors. Humans load blocks and supervise; the boom does the laying.
The honest limits: it’s built for low-rise structural walls (housing, low commercial), needs its specific block-and-adhesive system, and access to the service runs through FBR’s partner network. A skilled mason still wins on ornamental work, repairs, and anything bespoke — the robot wins on straight production walls, volume, and never taking a break.
### The competitive field
Construction Robotics’ SAM100 (~$500K) takes the collaborative approach — handling block placement while a human mason runs mortar and finishing. Netherlands-based Monumental raised $25M and completed a 15-meter commercial facade, betting on smaller, modular bricklaying robots. Different philosophies, same driver: housing demand is enormous and the masonry workforce is shrinking. For contractors, the practical takeaway is that robotic masonry is now a per-project service you can price, not a machine you must buy.
### FAQ
Q: How fast can a bricklaying robot work?
A: FBR’s Hadrian X lays up to roughly 360 large-format blocks per hour — and because the blocks are about twice standard size and the machine works long shifts, it can complete the walls of a house in about a day.
Q: How much does a bricklaying robot cost?
A: Hadrian X is offered as Walls-as-a-Service — contractors hire the system per project rather than purchasing it. The collaborative SAM100 from Construction Robotics costs approximately $500,000 to buy.
Q: Are robot-built brick walls as strong as traditional ones?
A: Hadrian X bonds blocks with a rapid construction adhesive instead of mortar, which FBR states is stronger than traditional mortar construction. The system has passed pilots under US building codes ahead of its 2026 commercial rollout.

## What construction robots cost — real numbers.
URL: https://whichbot.net/guides/construction-robot-cost
Target topics: construction robot cost, construction robotics ROI, how much do construction robots cost
Construction robotics pricing is refreshingly concrete compared to the consumer market — real list prices, real subscriptions, real payback data from deployed fleets. Here are the numbers that matter, in one place.
### The full cost stack nobody quotes upfront
Sticker prices miss four line items: payloads and attachments (a Spot with a laser scanner roughly doubles the base price), software subscriptions for mission planning and data processing, training and workflow setup (the robot is easy — changing your crews’ process is not), and BIM readiness (BIM-driven robots need clean models, and cleaning up your models has its own cost, though it pays off beyond robotics).
Reference points beyond our index: Built Robotics autonomous retrofits run roughly $150K–300K per machine; Construction Robotics’ SAM100 about $500K; DroneDeploy aerial survey from about $499/month. The pattern: data and layout tools are cheap and fast to repay; machines that physically build cost more and repay over years.
### The payback math that convinces CFOs
Three numbers close deals: rework (layout errors cause ~30% of it — a robot that eliminates layout error attacks the industry’s biggest waste stream), schedule (two weeks saved on a $50M project is $500K–$1M in carry costs), and safety (every fall or overhead-drilling injury avoided is direct cost plus insurance trajectory). Frame the robot against those, not against an hourly wage, and the ROI conversation gets short.
### FAQ
Q: How much do construction robots cost in 2026?
A: Representative real numbers: Boston Dynamics Spot from $74,500 plus payloads; Dusty Robotics FieldPrinter about $3,000–$5,000/month subscription; Construction Robotics SAM100 around $500,000; Built Robotics retrofits $150K–$300K; Ozmo and Hadrian X are quoted per building or per project as services.
Q: What is the payback period on construction robots?
A: Typical industry figures: inspection robots under 6 months, layout robots 6–12 months, bricklaying systems 3–5 years — with schedule compression (finishing weeks early saves carry costs) as the largest hidden return.

## Your building’s next hire is a robot.
URL: https://whichbot.net/guides/building-maintenance-robots
Target topics: building maintenance robots, facade robots, commercial building robots
Commercial buildings are quietly becoming robot employers. Facade cleaning, exterior coating, inspection walks — the dull, dangerous, at-height work that defines building maintenance is exactly what 2026 robotics does best. This guide is for owners, property managers, and facility contractors deciding where to start.
### Why at-height work automates first
Every task done from a swing stage, boom lift, or rope is expensive three times over: skilled labor is scarce (the window-cleaning workforce is aging out with few replacements), insurance for at-height work keeps climbing, and weather windows constrain scheduling. Robots attack all three at once — which is why facade work, not lobby cleaning, is where building robotics landed first. The pattern to expect: cleaning and coating now, facade inspection and repair detection next, minor facade repairs after that.
### A realistic adoption path for an owner
Start where you already spend: pull your facade cleaning and exterior painting contracts and get service quotes from robotic providers on the same scope — it’s a like-for-like comparison with zero capital. If you operate multiple properties, pilot on the building with the best roof rig and simplest facade, measure a full cycle, then roll out. Inspection robots come second, and make most sense if you already struggle to document conditions across a portfolio. None of this requires buying a robot — in 2026, building robotics is a procurement decision, not a capital one.
### FAQ
Q: What robots exist for commercial building maintenance?
A: Operational today: facade/window cleaning robots (Skyline Ozmo, deployed on Manhattan towers), large-scale exterior painting robots (PaintJet Bravo), and inspection/monitoring robots (Boston Dynamics Spot). All are available as services or purchases now.
Q: Do building maintenance robots require modifying the building?
A: Generally no — the leading systems are designed for existing infrastructure. Ozmo rides a building’s existing window-washing cradle (BMU), and PaintJet mounts to standard lifts already used on job sites.
Q: How do building owners get started with robots?
A: Request service quotes against your existing facade cleaning or painting contracts — the robotic providers price the same scope, making it a direct cost comparison with no capital purchase required.