Unitree
Unitree H2
- Autonomy
- IIAssisted
- Pre-release
- 75/100Promising
- Readiness
- Not assessed
- Price
- $29,900list
ComparisonUnitree H2 and EngineAI T800
EngineAI
EngineAI T800
- Autonomy
- IIAssisted
- Pre-release
- 61/100Plausible
- Readiness
- Not assessed
- Price
- $25,000list
Up to three
Autonomy Ladder
- IManual
- IIAssisted
- IIIConditional
- IVEnvironmental
- VGeneralized
All humanoid robots. Every figure below is placed against the whole class, so a gap can be read against how much the class itself varies.
Where each one suits you.
The conditions each robot was assessed against, side by side. Rows marked in gold are where these machines genuinely differ.
| Condition | Unitree H2 | EngineAI T800 |
|---|---|---|
| Where it operates | ||
| Home / residential | PoorPlatform targets industrial deployment only. No domestic autonomy capability disclosed. | PoorNo home-use documentation exists; 75 kg platform and $25K entry price preclude consumer home deployment. |
| Office / retail | DependsH2 is in pilot-phase industrial deployment. Commercial availability as a product remains unannounced. | DependsMax Edition targets 24/7 commercial deployment, but no published SLA or support structure is available to buyers outside China. |
| Warehouse / factory | GoodH2 demonstrates scripted task autonomy in factory environments with on-site safety supervision assumed. | GoodHigh-torque actuators and robust frame suit structured industrial pilot environments with on-site engineering staff. |
| Primary tasks | ||
| Household chores | PoorNo domestic task autonomy has been demonstrated or claimed. | PoorNo published household task capability; manipulation software for domestic chores is not documented. |
| Social / care | PoorH2 is positioned for task execution, not human interaction or companionship. | PoorNo interaction or communication features documented; positioned for physical tasks, not social engagement. |
| Specialized job | GoodDemonstrated in repetitive manufacturing tasks, picking, placement, line work. | GoodHigh-DOF body and optional dexterous hands suit specialized manipulation research and custom task development. |
| Supervision needs | ||
| Fully autonomous | PoorIndustrial deployments assume on-site operators and safety staff monitoring. | PoorAutonomous task completion without supervision is not documented; platform requires active monitoring in current form. |
| With oversightthey differ | GoodAutonomy is task-scripted; human oversight is integral to industrial pilot design. | DependsSupervised operation is feasible for research or pilot contexts, but software integration burden is high. |
| Teleoperated | GoodH2 is actively in pilot-phase factory deployments with structured handoff workflows. | GoodDesigned for pilot deployments with engineering support; first-batch buyers are expected to have integration capacity. |
| Timeline | ||
| Ready today | PoorH2 is pre-release. No shipping date or consumer pricing has been announced. | PoorConsumer-ready or production-ready deployment is not supported; platform is in early commercial rollout in China only. |
| Near-term (1–2 yrs) | DependsIf H2 commercializes within 12 months, availability will depend on Unitree's distribution strategy. | DependsInternational availability and task-level software maturity could unlock broader deployment within 12-24 months if software roadmap delivers. |
| Research phase | GoodH2 is currently a research and industrial pilot platform with demonstrated autonomy in controlled environments. | GoodCore positioning targets research institutions; ROS2 and NVIDIA compute stack support algorithm development workflows. |
A robot being poor for a condition is a fact about fit, not a fault. “Depends” means the record names a limit worth reading, not that the answer is unknown.
How much the claim can be trusted.
None of these has shipped, so they are assessed on track record, engineering plausibility and how the demonstrations match the claims.
Track Record35% of score
Has this maker shipped what it demoed before, on schedule, with the promised capability?
8558
- Unitree H285
- EngineAI T80058
28median 69 across 80 pre-release records100
Engineering35% of score
Are the demonstrated capabilities achievable in shipping form, or do they need breakthroughs not in evidence?
10071
- Unitree H2100
- EngineAI T80071
35median 70 across 87 pre-release records100
Demo Match30% of score
What has been shown to work, against what is claimed in marketing.
10054
- Unitree H2100
- EngineAI T80054
30median 66 across 87 pre-release records100
Published beside the number
Never averaged into it. These describe how much the maker has disclosed, not what the machine does.
| Unitree H2 | EngineAI T800 | |
|---|---|---|
| ReadinessShip-date specificity, pricing transparency, distribution and regulatory status. Published beside the number; never averaged in. | 25/100 | 70/100 |
| DisclosureHow much of the product is disclosed. Higher means fewer fundamental unknowns remain. Published beside the number; never averaged in. | 40/100 | 40/100 |
How far apart is a long way?
A number only means something next to the spread it sits in. Each track is the real range across every classified robot in the class, with the median marked.
Price18 with a published figure
$29,900$25,000
- Unitree H2$29,900
- EngineAI T800$25,000
$1,385median $22,500$250,000
Runtime34 with a published figure
180 min240 min
- Unitree H2180 min
- EngineAI T800240 min
60 minmedian 240 min720 min
Weight48 with a published figure
154.3 lbs165.3 lbs
- Unitree H2154.3 lbs
- EngineAI T800165.3 lbs
26.5 lbsmedian 142.6 lbs661.4 lbs
Height51 with a published figure
70.9 in68.1 in
- Unitree H270.9 in
- EngineAI T80068.1 in
34.7 inmedian 66.9 in86.6 in
Not compared, because fewer than two of these robots carry a published figure: Charge time, Noise.
What the numbers leave out.
Written on each record, and the only thing here every robot on the site carries.
The trade-off
What each design gives up to be good at what it is good at.
Unitree H2
Bipedal design demands precise balance control that manufacturing environments demand; humanoid hardware cost and complexity remain unpriced.
EngineAI T800
450 Nm peak torque and 14,000 W instantaneous joint power are genuinely high hardware specifications; the autonomy software layer to deploy that hardware on real tasks is undocumented.
Solid-state battery and modular hot-swap design solve a real operational problem. Four-edition tiering creates meaningful choice; Basic Edition lacks dexterous hands, limiting manipulation tasks entirely.
Who each one is wrong for
Stated by the record itself. A robot being wrong for a household is a fact about fit, not a fault.
Unitree H2
- Consumer home automation or domestic assistance
- Unitree publishes no domestic capability for H2: no cleaning, no fetching, no household task, and no autonomy that would run one. It is sold direct at $29,900 to anyone who orders it, so the barrier is capability, not availability.
- Any task that has to be completed unattended
- Unitree documents no autonomous task execution for H2, in structured environments or anywhere else. What is published is locomotion and whole-body control. Anything past that is software the buyer writes.
EngineAI T800
- Consumers seeking home utility without engineering background
- No home-use documentation, no consumer support channel, and $25,000+ pricing place this outside any current consumer deployment context.
- Organizations requiring plug-and-play task execution today
- Autonomy software capabilities are not documented at the task-completion level; the robot requires significant software integration before executing domain-specific workflows.
- International buyers outside China
- No official distributor, authorized reseller, or confirmed shipping route outside China has been announced as of June 2026.
- Operations needing documented uptime SLAs
- No published warranty, SLA, or commercial support structure available for the T800 as of evaluation date.
What stays your job
The upkeep ownership hands back to you, whatever the autonomy level says.
Unitree H2
- After every use: Inspect joints, shell panels and fasteners after dynamic-motion sessions; routines at that intensity load the frame far harder than walking does.
- After every use: Recharge between sessions. Unitree’s store page publishes no battery capacity or runtime figure for the H2, so session planning runs on observed runtime rather than a spec.
- As needed: Apply OTA updates. Unitree states the H2’s algorithms evolve continuously via OTA; no cadence or public changelog accompanies them.
- As needed: Actuator, joint or structural service through Unitree or an authorized reseller; no parts list or service interval is published for the H2.
- As needed: Consult Unitree’s H2 user manual for operating procedures; no maintenance schedule appears in Unitree’s accessible web material for the H2.
EngineAI T800
- After every use: Joint inspection for debris, actuator temperature check, and battery charge-level log.
- Weekly: Software update review and sensor calibration (LiDAR, stereo cameras, IMU).
- Monthly: Actuator lubrication and mechanical fastener torque inspection across 29 DOF.
- Every 3-6 months: Deep gait calibration, battery health cycle test, and dexterous hand tactile sensor verification.
- Annually: Full joint-module inspection and replacement of any wear components; service cadence not yet published by manufacturer.
- As needed: Firmware update deployment following EngineAI release notes; revalidate locomotion after each major update.
How much each of these is standing on.
Classifications rest on different depths of evidence. A pre-release assessment and a verified one are not the same kind of claim.
| Robot | Evidence status | Sources reviewed | Last assessed |
|---|---|---|---|
| Unitree H2 | Not Consumer-Ready | 3 | Sep 2026 |
| EngineAI T800 | Not Consumer-Ready | 8 | Sep 2026 |
Evidence status counts the sources behind a classification, not their kind. A high count can be maker documentation and price coverage, so read it as how much was reviewed rather than as independent testing. How we weigh evidence.
Keep going.
The full recordUnitree H2Specifications, evidence, the Fit Check in full, and every tracked change.
The full recordEngineAI T800Specifications, evidence, the Fit Check in full, and every tracked change.Written comparisonEngineAI T800 vs Unitree H2 vs Dobot Atom: what are they for?We have examined this pairing in full, with a stated finding.Robovations classifies robots; it does not rank them. Nothing on this page is a recommendation, and no manufacturer influences a classification or its placement here.
Add a robot
Comparing across classes is allowed — the page will switch to per-class positioning and tell you it has.