Unitree G1 vs 1X NEO: Price, Specs & Uses

Unitree G1 and 1X NEO are two of the few humanoid robots with public prices, but they are designed for fundamentally different buyers.

The Unitree G1 is a compact humanoid platform for robotics demonstrations, education and embodied-AI research. The US$13,500 base G1 is the lower-cost hardware, while the quote-based G1 EDU adds the secondary-development access, compute options and configurable hands that serious developers need. The 1X NEO is a full-size home robot offered through a US$20,000 Early Access purchase or a later US$499-per-month subscription, with domestic safety, voice interaction, household manipulation and a managed service at the centre of the product.

The short answer is clear: Unitree G1 is the better choice for laboratories, universities and developers—provided they buy G1 EDU rather than the locked base version. 1X NEO is the better choice for home use, human interaction, dexterous household manipulation and buyers who want a supported consumer experience rather than a robot to program.

Neither is a universal autonomous worker. G1 does not arrive as a ready-made housekeeper, and NEO’s first owners receive foundational autonomy with scheduled 1X Expert assistance for tasks the robot cannot yet complete independently.

Quick verdict: Choose Unitree G1 for the lowest acquisition price, compact transport and ready-made motion demonstrations. Choose G1 EDU for SDK access, ROS 2, low-level control, simulation and custom embodied-AI research. Choose 1X NEO if the robot will live around people and must prioritise soft construction, quiet operation, natural interaction, five-finger manipulation, self-charging and household services. Do not buy the base G1 expecting an open development platform, or preorder NEO expecting every advertised household task to be fully autonomous on day one.

Anton Robots lists the Unitree G1 and 1X NEO with their current prices, specifications and availability. Buyers can also open the direct Unitree G1 vs 1X NEO comparison for a side-by-side specification view.

Last reviewed: 15 July 2026. This comparison uses current manufacturer documentation and separates published hardware limits, included product functions and capabilities that depend on software, development or remote assistance. Specifications, delivery dates and features may change as both platforms evolve.

Unitree G1 vs 1X NEO at a Glance

The most important difference is not US$6,500. The base G1 price buys a non-secondary-development robot without dexterous hands. NEO’s Early Access price buys a home-focused hardware and service package, but not guaranteed full autonomy for every chore. Compare the configuration and experience you actually need—not the two headline prices in isolation.

Unitree G1 Models Compared

The name “Unitree G1” covers two materially different purchases. The base G1 is the US$13,500 version shown in Unitree’s online store. G1 EDU is the development platform, and its final specification and price depend on the selected joints, hands, wrist, waist and compute.

For buyers interested in motion demonstrations rather than development, the base Unitree G1 is a genuine lower-cost entry point. For an AI lab, university or startup, the EDU version is normally the meaningful configuration because the base product explicitly does not support secondary development.

Which G1 Is the Real Alternative to 1X NEO?

The base G1 is the price alternative to NEO; G1 EDU is the technical alternative. Neither is a direct home-use alternative.

• Base G1 vs NEO: the useful comparison for a buyer choosing between lower-cost humanoid hardware and a managed home robot. G1 is cheaper and available as hardware, but it is not open for secondary development and lacks NEO’s hands, domestic design and service model.
• G1 EDU vs NEO: the correct comparison for robotics capability. EDU supports custom control and compute, but its quote-based configured price can exceed the base headline substantially.
• G1 with Dex3-1 vs NEO: the relevant manipulation comparison. Unitree’s optional hand is a three-finger research end effector; NEO’s current shipping design uses two five-finger, high-DoF hands integrated with the complete home platform.

NEO is not the equivalent of a G1 EDU with friendlier covers. Its software, support, remote Expert Mode, app, charging behaviour and commercial terms form part of the product. Conversely, G1 EDU exposes a robotics-development stack that 1X does not currently offer to NEO buyers through a public SDK.

How We Compared the Robots

This comparison uses the current Unitree G1 specification table, Unitree’s store and official developer resources, plus the current NEO product page, order terms and 1X engineering updates.

We applied three evidence levels:

1. Published hardware specification: a measurable limit such as weight, battery capacity, arm load or IP rating.
2. Included product function: something the manufacturer says the buyer receives, such as G1’s built-in motions or NEO’s app and voice interface.
3. Demonstrated or developing capability: a task shown in a controlled video, enabled through custom development or expected to improve through later software.

Manufacturer demonstrations are not treated as proof that an untrained buyer can reproduce every task autonomously. No universal standard test currently allows G1 and NEO to be compared across household task success, fall rate, intervention frequency, grasp reliability or long-term uptime. This is therefore an evidence-based buying comparison, not an independent laboratory test.

Which Is Better: Unitree G1 or 1X NEO?

1X NEO is the better home robot. Unitree G1 EDU is the better developer robot. The base G1 is the cheaper but more limited purchase.

NEO wins when the robot must share living space with people. It is lighter despite being taller, has a soft and covered body, published whole-body ingress protection, quiet tendon actuation, substantially more capable hands, longer runtime, self-charging, natural-language interaction and a service path for tasks beyond its current autonomy.

G1 EDU wins when the buyer wants to build the intelligence rather than subscribe to the experience. Unitree provides SDK2, Python interfaces, ROS-compatible communication, simulation resources and low-level control. Its smaller body, LiDAR, quick-release battery and large research community make it a practical physical-AI platform.

The base G1 wins only a narrower contest: it is the least expensive way to acquire these two humanoid bodies and run supported built-in behaviours. Its US$13,500 price should not be presented as the cost of a fully equipped, programmable G1 with dexterous hands.

Price and Availability

Unitree G1 price

Unitree currently publishes the base G1 at US$13,500, excluding tax and shipping. The official Unitree store estimates shipping at approximately US$300–US$1,200, with the buyer responsible for customs duties, local taxes and import clearance. The product is currently marked as backordered, so “orderable” does not mean immediate dispatch.

G1 EDU is quote-based. The final cost depends on its joint configuration, dexterous hands, wrist and waist options, onboard compute, accessories, integration, freight and local support. Published reseller prices should be checked against the exact supplied bill of materials because two robots described as “G1 EDU” may not be equivalent.

1X NEO price

1X currently offers Early Access ownership for US$20,000, including a three-year warranty, premium support and priority delivery. A refundable US$200 deposit secures an order position. The Standard plan is listed at US$499 per month with a starter productivity package and standard delivery, but 1X says that subscription version will ship later.

The current order page states that US deliveries start in 2026. In an April 2026 production update, 1X said some customers would receive NEO during the year and others later, after its initial annual production capacity was booked rapidly. A preorder is therefore a place in a delivery programme, not the same as buying in-stock hardware.

Price winner

The base G1 has the lower purchase price, but there is no honest universal price winner. A developer needs a G1 EDU quote. A home buyer needs to compare NEO’s purchase or subscription terms with the support, updates and Expert services included. The relevant number is the delivered configuration over the intended ownership period.

Size, Weight and Transport

G1 stands 1.32 m tall, measures 450 mm wide and 200 mm thick, and folds to 690 × 450 × 300 mm. It weighs about 35 kg with its battery. Its folding geometry is valuable for a laboratory that moves the robot between a protective rig, vehicle, classroom and demonstration site.

NEO is human-scale at 5 ft 6 in, or approximately 1.68 m, yet weighs only 66 lb or 29.9 kg. The extra height provides access to worktops, shelves, doors and appliances designed around adult human reach. The lower mass supports domestic safety, although it remains a large powered machine that requires clear operating space.

Size verdict: G1 is easier to store and transport. NEO offers more useful household reach while weighing about 5 kg less.

Mobility, Speed and Balance

Unitree built its reputation around dynamic legged motion, and G1 is the more established research platform for locomotion. Unitree states a trotting speed above 2 m/s and demonstrates standing up, sitting, folding and highly dynamic learned behaviours. G1 EDU allows teams to work closer to the control layer through SDK and simulation resources.

1X publishes a normal walking speed of 1.4 m/s and a maximum running speed of 6.2 m/s. Its Redwood mobility controller has demonstrated multidirectional walking, sitting, standing, kneeling, rising from the floor and stair use with stereo vision. Those figures describe the platform’s mobility envelope; they do not mean a first-delivery NEO will run through an occupied home at maximum speed or autonomously negotiate every staircase.

For practical buying decisions:

• Locomotion research and custom control: G1 EDU
• Compact, dynamic demonstrations: G1
• Human-scale household movement: NEO
• Published maximum running speed: NEO
• Transparent developer path for new locomotion policies: G1 EDU

Hands, Manipulation and Strength

This is the largest hardware difference between the robots.

The base G1 does not include a dexterous hand. G1 EDU can be configured with Unitree’s Dex3-1, a three-finger force-controlled hand with seven active DoF, optional tactile arrays and a specified 500 g grasp under Unitree’s test orientation. The G1 body has a stated maximum arm load of about 2 kg on the base model and 3 kg on EDU, and Unitree warns that load varies substantially with arm extension.

NEO uses two integrated five-finger tendon-driven hands. 1X’s July 2026 shipping-hand announcement describes 25 DoF per hand: 22 fully actuated DoF across the fingers and palm plus three at the wrist. The joints are force-controlled and backdrivable, with tactile and shear sensing across the hand, IP68 sealing and stated ±0.2 mm positioning accuracy. 1X publishes an 18 lb or approximately 8.2 kg arm payload, a 55 lb or 25 kg carrying figure and a 154 lb or 70 kg whole-body lift figure.

Arm payload, carrying capacity and whole-body lift are not interchangeable. They describe different postures and load paths. Neither the 70 kg NEO lift figure nor G1’s maximum arm load proves that the robot can safely identify, grasp and autonomously move an arbitrary object in a real home or workplace.

NEO is the clear winner for included manipulation hardware and human-shaped task coverage. G1 EDU remains useful when the research question requires a configurable end effector, direct control or a lower-cost experimental body rather than the most capable included hands.

Perception and Onboard Compute

G1 combines a depth camera with 3D LiDAR, giving developers geometric sensing around the robot and a practical starting point for mapping, navigation and embodied perception. Both G1 versions include an eight-core CPU, while EDU can be ordered with additional compute such as NVIDIA Orin modules.

NEO takes a vision-first approach. Its current specification lists dual 8.85 MP fisheye cameras operating at 90 Hz, four beamforming microphones, link-level IMUs, Wi-Fi, Bluetooth and 5G. The 1X NEO Cortex uses NVIDIA Jetson Thor with up to 2,070 FP4 TFLOPS of stated AI compute. This is a far stronger included inference platform on paper than G1’s base CPU.

The sensors reflect the products’ priorities. G1’s LiDAR is valuable for robotics research and spatial mapping. NEO’s stereo vision, audio array and high AI compute are designed for visual-language interaction and manipulation in human environments. More compute does not itself prove higher task success; the model, data, control policy, latency and safety layer determine how effectively it is used.

AI, Autonomy and Expert Mode

Both manufacturers use ambitious AI language, but the delivered experiences are different.

Unitree positions G1 around imitation learning, reinforcement learning and its UnifoLM robot model. The robot can run supported motion behaviours and receive OTA updates. G1 EDU gives a capable team the hardware and interfaces to build perception, navigation, locomotion and manipulation systems. It does not turn those research ingredients into an out-of-the-box general worker.

NEO uses 1X’s Redwood AI to learn and repeat household tasks. Its consumer layer includes a built-in language model, voice interface, visual and audio context, memory, a mobile app and chore scheduling. 1X says Early Access robots arrive with foundational autonomy and grow through continued use and software updates.

For a chore NEO does not know, the owner can schedule Expert Mode. A 1X Expert remotely supervises or guides the robot so the task can be completed and provide learning data. This is a legitimate route to usefulness, but it must not be described as autonomous task completion.

AI verdict: NEO offers the more complete AI product experience. G1 EDU offers the more open environment for teams developing their own physical-AI systems.

SDK, ROS 2 and Development Access

G1 EDU is the decisive winner for developers. Unitree provides official SDK2 and Python interfaces, DDS-based communication, low-level commands and state access, G1 models, and MuJoCo simulation designed for sim-to-real controller work. The broader Unitree ecosystem also includes ROS resources and a large community building policies and integrations around G1.

The distinction between variants is non-negotiable:

• Base G1: no secondary development according to Unitree’s current product table and store.
• G1 EDU: intended for secondary development, custom compute, low-level control and advanced robotics work.

NEO’s current public product and order documentation does not present a buyer-facing SDK, ROS 2 package or low-level control path. The platform contains exceptional hardware and 1X discusses what developers could build conceptually, but a consumer should not infer public programmability from that language. Ask 1X directly about APIs, local integrations, data export and third-party development before purchasing for a software project.

Development verdict: Choose G1 EDU for a robot you intend to program. Choose NEO for a robot whose managed software experience you intend to use.

Battery Life and Charging

Both G1 versions use a quick-release 9,000 mAh battery and have a stated runtime of about two hours. A removable battery can reduce downtime when a team owns charged spares, but battery replacement and charging remain operator-managed.

NEO uses an 842 Wh battery with a stated four-hour runtime. It manages its own charge level and physically plugs itself into the supplied charger. 1X also states that quick charging can add one hour of runtime in six minutes.

Actual runtime depends on walking, lifting, manipulation, compute load, radio use, floor conditions, temperature and standby time. NEO wins on published endurance and autonomous charging; G1’s advantage is a quick-release pack that suits supervised lab workflows.

Safety, Noise and Durability

G1 is compact but not designed like a domestic appliance. Its rigid structure, high-torque joints and dynamic motions can injure people or damage the robot in a fall. Unitree explicitly warns users to keep sufficient distance and operate with caution. The current specification table does not publish a whole-robot IP rating, so buyers should not assume G1 can operate in rain, wet floors or dusty conditions.

NEO was engineered around close human contact. Its hardware is wrapped in a soft custom lattice polymer, its joints are externally covered to remove accessible pinch points, and its low-inertia tendon drives are highly backdrivable. 1X publishes a Head Injury Criterion below 250, a 22 dB noise figure, IP44 protection for the body and IP68 for the hands. The suit and shoes are removable and machine washable.

IP44 means protection against splashing water, not submersion, pressure washing or unrestricted outdoor use. IP68 applies to NEO’s hands, not the entire robot. Soft construction also reduces risk rather than eliminating it; payloads, falls, hot objects, stairs, pets and children still require a controlled operating policy.

Domestic safety winner: 1X NEO.

Privacy, Connectivity and Remote Operation

NEO’s usefulness depends partly on cameras, microphones, memory, cloud-connected services, remote control and optional access by a 1X Expert. 1X says Expert assistance is scheduled for tasks the robot does not know. Prospective owners should still understand exactly when video and audio leave the robot, how remote access is indicated and authorised, what data is retained, how training permissions work, and which features remain available if connectivity fails.

G1 is normally operated inside a lab or controlled site by its owner’s team. Its privacy questions are therefore different: network isolation, developer credentials, telemetry, OTA updates, camera and LiDAR data, and access to the added compute module. An organisation deploying either robot should complete a security and data-governance review rather than treating it like an offline mechanical tool.

Privacy verdict: G1 EDU gives the owner more control over a custom development environment. NEO provides a more convenient remote service, but that convenience requires a more careful review of access and data terms.

Can Either Robot Work Autonomously?

Yes, within narrow and changing limits—but neither should be described as fully autonomous in the general human sense.

G1 can balance, move and run supported behaviours, and EDU teams can build autonomous functions around its sensors and control interfaces. Task-level autonomy is primarily a development project. A video of G1 dancing, boxing or recovering from a disturbance proves a particular motion policy, not general household reasoning or industrial productivity.

NEO is designed to navigate a home, respond to natural-language requests and perform a growing set of chores. 1X openly states that early owners receive basic autonomy and can schedule Expert assistance for unknown tasks. NEO therefore has a better route to practical home task completion, but some of that outcome may involve human support rather than the robot acting alone.

The right procurement questions are:

• Which exact tasks are autonomous at delivery?
• What percentage of attempts complete without intervention?
• What objects, surfaces and room layouts were validated?
• How does the robot fail safely?
• What happens without internet access?
• Which tasks require Expert Mode, owner teleoperation or custom code?

Best Robot by Use Case

Explore more humanoid robots for sale, home robots and educational robots if neither platform matches the application.

Unitree G1 Pros and Cons

Pros

• US$13,500 published base price
• Orderable hardware rather than a future-only concept
• Compact 1.32 m body that folds for transport
• Depth camera and 3D LiDAR included
• Strong dynamic motion and locomotion research value
• G1 EDU supports secondary development and additional compute
• Official SDK2, Python and simulation resources
• Quick-release battery
• Growing research and developer community

Cons

• Base G1 does not support secondary development
• US$13,500 configuration has no dexterous hands
• G1 EDU price is not public and varies by hardware
• Shorter published runtime than NEO
• Lower arm-load figures
• No published whole-robot IP rating
• Rigid, powerful platform not designed for unsupervised domestic use
• No complete consumer chore, app and managed-assistance experience equivalent to NEO

1X NEO Pros and Cons

Pros

• Designed specifically for living and working around people at home
• Human-scale reach at a relatively low 29.9 kg body weight
• Two five-finger force-controlled and tactile hands
• Much higher published arm payload than G1
• Four-hour stated runtime and self-charging
• Soft body, covered joints and low-inertia tendon drives
• IP44 body and IP68 hands
• 22 dB stated noise level
• Jetson Thor compute and a vision-language household AI stack
• Voice, app, memory, remote control and scheduled Expert Mode
• Three-year warranty with Early Access ownership

Cons

• Preorder and early-access product rather than generally available stock
• US$20,000 purchase price or US$499 monthly subscription
• First deliveries begin in 2026, but individual timing is not guaranteed by the headline
• Early owners receive foundational rather than universal autonomy
• Some chores may require scheduled remote Expert assistance
• No public buyer SDK or ROS 2 development path documented on current product pages
• Camera, microphone, memory and remote access require careful privacy review
• IP44 body is splash-resistant, not waterproof or outdoor-industrial protection
• Published maximums do not yet provide independent long-term household reliability data

Total Cost of Ownership

G1 costs should include the correct model, hands or end effectors, extra compute, spare batteries, charger, controller, freight, customs, tax, safety rig, protective area, replacement parts, engineering time and support. The cheapest G1 becomes the wrong purchase if a team later discovers that secondary development requires EDU.

NEO costs should include the Early Access purchase or subscription, service coverage after the included period, connectivity, accessories, household preparation and the commercial terms around Expert Mode and future productivity packages. Buyers should confirm whether ownership requires any continuing software or service payment for the functions they consider essential.

The economic case is also different. G1 EDU is justified by experiments, publications, student access, datasets, demonstrations or product development. NEO is justified by useful household hours, accessibility, telepresence and time saved. Compare measurable outputs rather than attempting to calculate value from hardware specifications alone.

G1 is a capital purchase plus engineering programme. NEO is a consumer hardware-and-service relationship.

Who Should Buy a Unitree G1?

Choose G1 if most of the following are true:

• You need physical humanoid hardware now rather than a place in a preorder queue
• Your primary use is education, demonstrations or robotics R&D
• You value compact storage and transport
• LiDAR and depth sensing fit your research
• You have trained operators and a controlled testing space
• You can define whether built-in motion or secondary development is required before purchase
• You are prepared to select and integrate hands, compute and software

Choose the base model only for supported behaviours and demonstrations. Choose G1 EDU if you need SDK access, ROS 2, low-level control, custom AI, hands or research compute. Review the complete Unitree G1 price, specifications and availability before requesting the exact configuration.

Who Should Buy a 1X NEO?

Choose NEO if most of the following are true:

• The robot is intended primarily for the home
• Human-scale reach and five-finger manipulation matter
• You prioritise soft construction, low noise and close interaction
• You want voice commands, chore scheduling, self-charging and app control
• You accept that Early Access autonomy will expand over time
• You are comfortable using scheduled Expert assistance for unsupported chores
• You have reviewed the privacy, connectivity and remote-access terms
• You are located in a supported delivery market and can tolerate uncertain individual timing

Review the current 1X NEO price, specifications and availability and confirm delivery eligibility before placing a deposit.

Common Buying Mistakes

• Comparing US$13,500 with US$20,000 as equivalent packages: the base G1 lacks secondary development and dexterous hands.
• Buying base G1 for SDK work: Unitree explicitly directs development buyers to G1 EDU.
• Assuming every G1 EDU is identical: DoF, wrists, waist, hands and compute vary by configuration.
• Treating NEO’s US$200 deposit as its price: it reserves an order; ownership is US$20,000 under the current Early Access offer.
• Assuming NEO is fully autonomous at delivery: 1X says early robots begin with foundational autonomy and use Expert Mode for unknown tasks.
• Confusing demonstrated dexterity with autonomous task success: capable hands still need perception, planning and a reliable policy.
• Confusing lift, carry and arm payload: these numbers describe different mechanical conditions.
• Assuming NEO is an open development robot: verify APIs and SDK access in writing.
• Assuming either robot is waterproof: NEO’s body is IP44 and G1 has no published whole-body IP rating.
• Ignoring privacy and security: both robots use cameras, microphones, networks and software updates; NEO also offers remote Expert access.
• Buying before defining success: specify the task, environment, intervention rate, safety case, data output and support requirement first.

FAQs

Is Unitree G1 better than 1X NEO?

G1 EDU is better for robotics development, ROS 2, low-level control, locomotion research and custom embodied AI. NEO is better for home use, natural interaction, safety around people, five-finger manipulation, battery life and managed household assistance.

Which is better for home use, Unitree G1 or 1X NEO?

1X NEO. It was designed for homes with a soft body, covered joints, 22 dB operation, IP44 body protection, self-charging, voice control, chore scheduling and Expert assistance. G1 is a research and demonstration platform, not a ready-made domestic helper.

Which is better for research?

Unitree G1 EDU. It supports secondary development, official SDK resources, low-level commands, simulation and configurable compute. The base G1 is not the correct model for advanced development.

How much does Unitree G1 cost?

The base Unitree G1 is currently listed at US$13,500 before tax and shipping. G1 EDU is quote-based because its hands, joints, compute and other options vary.

How much does 1X NEO cost?

1X lists Early Access ownership at US$20,000 and a Standard subscription at US$499 per month. A refundable US$200 deposit reserves an order. The subscription version is scheduled to ship later than priority Early Access units.

Is the US$200 NEO deposit refundable?

The current 1X order page describes the US$200 deposit as fully refundable. Buyers should still read the terms shown at checkout for their order and country.

Can you buy Unitree G1 now?

Unitree accepts orders for the base G1, but the official store currently marks it as backordered. Delivery time, freight, customs and local support should be confirmed before payment.

When will 1X NEO ship?

1X says US deliveries start in 2026. The company has also said some customers will receive robots in 2026 and others later as production and home testing scale. Confirm the expected delivery window for your position rather than relying only on the launch year.

Does Unitree G1 come with hands?

The US$13,500 base G1 specification does not include dexterous hands. G1 EDU can be configured with Unitree’s three-finger Dex3-1 or other compatible end effectors.

How many degrees of freedom does 1X NEO have?

1X’s July 2026 hand announcement states 25 DoF per hand, comprising 22 across the fingers and palm plus three at the wrist. The current product page separately lists seven per arm, three in the neck, two in the spine and six per leg. Because the wrist may overlap the hand and arm descriptions, buyers should not add every figure into a single total without a complete joint map from 1X.

Which robot has better hands?

NEO has the stronger included hand hardware. It uses two five-finger, force-controlled, tactile and IP68-sealed hands. G1’s optional Dex3-1 is a useful three-finger research hand, but it has fewer DoF and a much lower specified grasp load.

Which robot is stronger?

NEO has the higher published manipulation figures: an 18 lb arm payload, 55 lb carrying figure and 154 lb whole-body lift. G1 lists about 2 kg maximum arm load for base and 3 kg for EDU. These measurements are not identical tests, so validate the exact task and posture.

Which robot is faster?

NEO has the higher published maximum at 6.2 m/s, while G1 has a stated trotting speed above 2 m/s. Maximum running speed is not a measure of safe autonomous speed in a home, laboratory or workplace.

Which robot has better battery life?

NEO. It has a stated runtime of four hours compared with about two hours for G1. NEO also self-plugs into its charger, while G1 uses a quick-release battery and operator-managed charging.

Can Unitree G1 do household chores?

A G1 EDU can be developed for selected manipulation and navigation tasks, but it does not arrive as a general home assistant. The base G1 lacks secondary-development access and dexterous hands, making it particularly unsuitable for that expectation.

Can 1X NEO fold laundry?

1X presents laundry handling as a target chore and NEO’s hands are designed for deformable objects. Early owners should expect task performance to depend on the item, setup, software version and whether scheduled Expert Mode is required. It is not evidence that every garment can be folded autonomously on day one.

Is 1X NEO fully autonomous?

No. 1X says NEO is built toward full autonomy but Early Access units arrive with foundational autonomy. Tasks the robot does not know can use scheduled remote Expert assistance.

Can you program Unitree G1?

Yes, if you buy G1 EDU. Unitree’s current store states that the base G1 does not support secondary development and directs customisation buyers to the EDU edition.

Does Unitree G1 support ROS 2?

The Unitree development ecosystem uses DDS communication compatible with ROS-style workflows and provides official SDK, Python, simulation and robot-description resources. Purchase G1 EDU and confirm the current supported packages, operating system and interfaces for the intended project.

Can you program 1X NEO?

1X does not currently document a public buyer-facing NEO SDK, ROS 2 package or low-level control interface on its product and order pages. Ask 1X for written API and developer-access details before buying NEO for custom software development.

Is Unitree G1 safe around children?

G1 is a powerful rigid humanoid, and Unitree tells users to maintain a sufficient safety distance. It should be operated by trained adults in a controlled area, not treated as an unsupervised child-safe appliance.

Is 1X NEO safe around children?

NEO has a much stronger domestic safety design than G1, including soft covers, inaccessible pinch points, low-inertia tendon drives and a stated HIC below 250. That does not remove the need for supervision, safe task rules and confirmation of 1X’s current guidance for children and pets.

Is Unitree G1 waterproof?

Unitree does not publish a whole-robot ingress-protection rating in the current G1 specification table. Treat it as a controlled indoor platform unless the supplier provides written environmental limits for the exact configuration.

Is 1X NEO waterproof?

NEO’s hands are IP68 and its body is IP44. The hands can tolerate immersion under the specified conditions, while the body is protected against splashing water. The complete robot is not waterproof or intended for submersion.

Does NEO require a human operator?

Not for every function. NEO has foundational autonomous capabilities, but difficult or unknown chores can be scheduled for a 1X Expert to guide remotely. Owners can also pilot NEO through the mobile app and VR.

Which robot is easier to buy?

G1 is easier to purchase as current hardware because Unitree accepts online orders, although it is backordered and international buyers must manage freight and import requirements. NEO currently requires a preorder and supported delivery location.

Which should I buy: Unitree G1 or 1X NEO?

Buy the base G1 for affordable humanoid hardware and demonstrations. Buy G1 EDU for robotics development, ROS 2, simulation, low-level control and custom embodied AI. Buy NEO for home assistance, human interaction, five-finger manipulation and a managed consumer experience—while accepting its early-access autonomy and delivery constraints.

Final Verdict

Unitree G1 and 1X NEO do not compete for the same job. G1 is the better robot to build on; NEO is the better robot to live with.

The base G1 deserves attention because it makes capable humanoid hardware orderable at US$13,500. Its compact folding body, dynamic motion, depth camera and 3D LiDAR are compelling for demonstrations and education. But the headline configuration is not the programmable, dexterous research platform many comparisons imply. Serious development requires G1 EDU and an exact quote.

NEO costs more because it combines hardware with a domestic product layer. It is taller but lighter, quieter, softer, stronger at the arms, far more dexterous at the hands and designed to interact through voice, vision, an app and a managed service. Its four-hour runtime and self-charging also make far more sense in a home. The trade-off is that it remains an Early Access preorder whose autonomy will grow over time and whose Expert Mode can involve remote human assistance.

Start with the outcome. If you want to create and test robot behaviour, specify a G1 EDU. If you want help around the home, evaluate NEO’s current task list, intervention requirements, privacy terms and delivery window. Then compare Unitree G1 and 1X NEO side by side or use the Anton Robots finder to match a robot to your task, environment and budget.

Primary Sources

• Unitree G1 official product page and specification table
• Unitree official G1 store, pricing, shipping and development notice
• Unitree G1 developer documentation
• Official Unitree SDK2 repository
• Official Unitree SDK2 Python repository
• Official Unitree MuJoCo simulation repository
• Unitree Dex3-1 official specifications
• 1X NEO official product page and hardware specifications
• 1X NEO official pricing, order and Early Access information
• 1X NEO launch, capabilities and availability announcement
• 1X NEO production and delivery update
• 1X NEO 25-DoF hands engineering specifications
• 1X Redwood mobility controller and demonstrated movements