RELEASED 2025REASSESSED 2026.09.02

Ropet

Ropet KAMOMO

Name: Ropet KAMOMO
Brand: Ropet

Ropet KAMOMO reacts to voice, touch, and gesture with animated responses that the manufacturer states are processed locally by default, with one optional cloud-connected feature. It does not navigate or act independently, and its published pricing and physical specifications are not consistently documented by the manufacturer.

Autonomy Ladder

Why Level I.

One five-rung scale, every category, from acting only on command to working unsupervised anywhere.

VGeneralized AutonomyAny task, any setting. No consumer robot has reached it.
IVEnvironmental AutonomyCopes with everyday clutter and frees itself, shown by independent testing.
IIIConditional AutonomyFinishes whole jobs in familiar spaces; struggles at the edges.
IIAssisted AutonomyRuns preset routines; stalls at anything new.
To reach Level IIRuns a scheduled routine end to end without a command for each step.
IManual AutomationThis robot

Ropet KAMOMO acts only on command, one motion at a time.
What it does on its own
- Responds to voice, touch, and gesture with animated reactions; recognizes up to 100 faces per the manufacturer; develops one of four selectable personality types through repeated interaction; processes camera and microphone input locally by default and can function fully offline, per the manufacturer; supports interchangeable fur coats, masks, and eye designs.
Why this level
- The device reacts to voice, touch, and gesture with animated responses, and the manufacturer states its 'personality' evolves through repeated interaction while default camera/microphone processing runs fully on-device.
- It does not navigate, move through a room, or initiate actions on its own.
- A human charges it, selects its personality and customization, and manually enables optional features such as Dream Sketch.
What holds it at Level I
- Does not move through or navigate a room; all interaction is triggered by direct human input rather than self-initiated action. The manufacturer's product page does not publish physical dimensions or weight. A stable manufacturer's suggested retail price could not be confirmed: the site displays inconsistent regular-price anchors ($469 vs. $489) alongside current sale pricing, and CES 2025 debut pricing ($299 base / $319 Pro) differs from both.
4 sources reviewed

How we assign a level →

Human Readiness

Ready Now.

One question: what does buying this robot today walk you into? The state is computed from the record, and recomputed daily.

Ready NowThis robot

Ropet KAMOMO is on sale, assessed, and nothing on its record argues for waiting.
How this was derived
- Shipping past the first-90-day ownership window with a published assessment, no documented issues on record, and no announced successor.
Notes on the record
- KAMOMO now sells through multiple established retail channels with published dimensions, weight, and a price band that is consistent across configurations, well past the launch window that justified caution.
- No successor, safety incident, or independent report contradicting its stated companion function turned up, so the ordinary consumer gates apply.
On the record
- Shipping since January 2025 · 20 months on the market
- Retail listings on file: Amazon
- No successor announced
- 4 sources reviewed
What would move it
A documented problem, an announced successor, or a withdrawn listing moves it out; the rating recomputes daily.
New to MarketOn sale under 90 days; the ownership record is still forming.
WaitA successor is announced and has not shipped; re-read the decision when it arrives.
Known IssuesDocumented problems stand on the record; read the case before buying.

How we assess readiness →

Robovations Score

How this number was reached

47of 100

LimitedAt or above 4% of the 436 robots we scoreand 8% of the 24 scored specialty robots

DeficientLimitedCapableStrongExemplary

Measured on 7 of the 9 inputs26% not counted

Our assessment did not establish an independent source for edge-case recovery, field failure rate on this machine, so those inputs are left out of the score rather than guessed at, and the rest re-weighted to fill the space. A gap in what we have been able to verify is not evidence against the robot. Where a whole pillar has nothing measured, we publish no number at all.

What moved itpoints gained or lost against a typical robot

Held back by

Task completion52−3.2
Software stability45−2.8
Support longevity35−2.7

Capability40% of the score50What it does without you.

Task completion16% of the score52

Evidence

A backer who received a Ropet Pro in Brisbane on 29 August 2025 reports it working immediately: 'Ropet works right out of the box', with the interaction carried by vocalizations and eye behavior rather than movement through a space. Their summary of what it delivers is measured: 'Although it might not seem to do much, I feel for the price it packs a lot of functionality.' Like Moflin it is stationary and owner-initiated, and the owner performs the charging, so the self-service limb of this referent is not met. Scored at the point where discrete portions of the interaction work reliably and a human routinely closes the loop.

Coverage and navigation10% of the score48

Evidence

Attentional coverage from a fixed position, which 4A expressly permits, and the sensing quality is the record's strongest documented feature. The same owner account describes eye tracking as 'incredibly precise', with pupils that 'dilate after blinking and move much more dynamically than I realised', and notes that fur covering the body makes petting register across the whole surface rather than at designated pads. What is not evidenced is any person model: nothing establishes that it distinguishes or remembers individuals, so coverage is of whoever is in front of it.

Edge-case recoverynot countedno source verified45

Evidence

Unevidenced. Recovering from a lost or refused interaction and re-engaging appropriately is the behavior this input measures, and no source read for this assessment observes it. The available first-hand material is an unboxing and first-impressions account written the day the unit arrived, which explicitly had not yet reached the app, and no longitudinal owner reporting exists for a product this new. Excluded from the composite rather than scored from the vendor's description of its own behavior model.

Operating envelope4% of the score45

Evidence

Thin and scored conservatively. Ropet is a desk or lap device that depends on its camera, so light level is a real condition, and it has no locomotion with which to leave an unfavorable position. Household variety, multiple people, noise and routine change, is the substance of this input for a companion and none of it is documented for this platform. Placed at the constrained-setting anchor on what the hardware plainly implies, with the absence of evidence stated rather than resolved in the product's favor.

Dependability40% of the score41Whether it keeps doing it.

Field failure ratenot countedno source verified55

Evidence

First units reached testers in May 2025, and general backer shipments ran through that summer. No recall, no regulatory record and no documented failure mode surfaced. No public owner channel of any size exists for this platform, and the available reporting is backer posts and vendor material. Section 4.5 is explicit that absence of a failure record is absence of evidence, not absence of failure. No independent source reports on failures for this machine, so the input is left out of the score.

Software stability14% of the score45

Evidence

Updates ship and have already regressed shipped behavior, which is the 50 anchor's second limb. An OTA channel is active, a developer interview noting firmware updates large enough that owners on metered connections have asked for differential downloads, and the vendor pointed to a major release for February 2026. Against that, owner reporting records that 'some functions stopped working after the update' and that language settings were 'reset once'. No release notes naming the defect fixed or the capability changed were found anywhere public, so a buyer cannot inspect what any release did. Scored just below the anchor because the regression is documented outside notes that do not exist, which is the weaker of the two states that anchor describes.

Support longevity10% of the score35

Evidence

A first-time vendor with no conduct record at all, shipping its first product out of a crowdfunding campaign, and section 4.7 is explicit that a vendor with no support history has not demonstrated the 75 anchor and does not satisfy it by absence. Conversation depends on a cloud LLM service with no stated term or remedy, which is the 50 anchor's condition, and it is scored below that because there is no parts channel, no published support horizon, and no generation of evidence about whether this organization sustains a line. Nothing here is a finding against the product; it is the honest position of a platform with no history yet.

Ownership burden20% of the score52What it asks of you.

Routine maintenance12% of the score50

Evidence

Light in absolute terms and owner-carried throughout. Ropet is small, stationary and charged by hand, so the owner is in the power loop for every cycle, and the fur body invites the same periodic cleaning that Moflin's does. There is no dock behavior to fail and nothing to disassemble. Scored at the midpoint: each action is trivial and tool-free, but the frequency is set by the owner rather than by the machine, against an engagement output that is modest.

Consumables8% of the score55

Evidence

No subscription is documented, the conversational layer is included rather than gated behind a recurring fee, and there are no filters, pads or media to buy. Under this pillar's normalization rule the absence earns moderately rather than highly, because it follows in part from a device that delivers a modest output, which is the same reasoning that holds Aibo at 47 and EBO X at 47. No wear item requiring scheduled replacement is documented, though the field record is too short for one to have emerged.

middle half of the databaseits medianthis robot

Reported, not scoredneither measures how well the robot works

Data postureCloud-optional

Full core function without an account; cloud adds convenience.

What leaves the homeDocumented

The device has both a camera and microphone. The manufacturer states all image and audio data are processed locally on-device by default and that the unit can function fully offline. The optional Dream Sketch feature, which must be manually enabled, encrypts photos on-device before uploading them to a cloud AI service, per the manufacturer.

On-device processingYes

Read the full methodology →

FIT CHECK

Will the Ropet KAMOMO fit your application?

Describe your use case using the controls. Each choice reads the spec sheet for you, flagging where the Ropet KAMOMO is built for your space and where it isn’t.

Room Lighting

Cloud Feature Comfort

Mobility Expectation

Engagement Style

A classification of fit for the conditions you set, not a recommendation. Verdicts describe engineering design intent, not test results.

SPECIFICATION

The full spec sheet.

Manufacturer-published figures, measured against every specialty robot we track.

Dimensions not publishedDimensional drawing

Key specificationsManufacturer-published

Charge time: Approximately 1-2 hours
Face recognition capacity: Up to 100 faces, per manufacturer

Where it lands · ranked across 44 specialty robotsThis robotMedianTypical range

Runtime210min

30 min960 min

Near median

Position shows where the figure sits across the tracked category, not a quality judgment.

Official referencesManufacturer support Product page

OWNERSHIP

Living with it.

Plush companion with no published field record; battery endurance is its one documented limit.

No independent reliability data exists. Ropet documents 2.5-3.5 hour battery life, a one-year warranty, and manual charging on a base; its help FAQ covers fur and mask care.

Owner effort~15 minhands-on per month

ConsumablesNonereplaceable parts / yr

Reliability trendUnprovenowner-reported arc

What goes wrong · 4 documented · 0 self-recover

Common

Battery depletion mid-sessionInteraction stops until recharged

Owner fixReturn to charging base; about 1-2 hours

Occasional

App pairing failuresCustomization and updates blocked

Owner fixRe-pair per Ropet's FAQ steps

Occasional

Fur jacket soilingHygiene, worn appearance

Owner fixClean or replace jacket per care FAQ

Rare

Cloud feature outageDream Sketch unavailable

Owner fixCore functions continue locally; contact support

Upkeep routine

Each useEvery cycle

Return KAMOMO to its charging base; a full charge lasts about 2.5-3.5 hours of interaction and takes about 1 hour fast or 1.5-2 hours standard, per Ropet.

As neededno fixed schedule

Clean or swap the fur jacket and masks; Ropet's care FAQ covers mask replacement and fur jacket cleaning but publishes no cadence. · Wipe the camera window and touch surfaces gently when dusty; touch sensors sit beneath the fur. · Install app and firmware updates; Ropet documents releases through its app update notes. · Re-enable optional cloud features such as Dream Sketch manually; photos are encrypted before leaving the device, per Ropet.

Safety notes

NoteLocal processing by default: Ropet states image and audio are processed on-device and it functions fully offline.

NoteOptional cloud photo feature: Dream Sketch must be enabled manually; photos are encrypted before leaving the device.

NoteBody-heat element runs warm: Ropet's product copy describes a constant 39 C body warmth from a mild heating element.

RECORD

How the assessment has moved.

Every announcement, release and reassessment behind the current classification.

Coming soon

Just launched

Active updates

Settled

Discontinued

Sep 2026 · AssessmentRobovations Score revised 49 → 47
Sep 2026 · AssessmentRobovations Score rescored under v4.2
Jan 2025 · DemoCompanion robot draws press coverage at CES 2025
Dec 2024 · AnnouncementLaunches on Kickstarter with $169 early-bird pricing

Common questions

What people actually ask.

5 answered

OperationDoes Ropet KAMOMO need an internet connection to work?

No. The manufacturer states the device processes camera and microphone data locally by default and can function fully offline; only the optional Dream Sketch feature uses a cloud AI service, and it must be manually enabled.

OperationCan it move around a room on its own?

No. It is a handheld/lap-sized companion device that reacts to voice, touch, and gesture rather than navigating or self-locomoting.

OperationHow long does the battery last?

The manufacturer states approximately 2.5 to 3.5 hours of use per full charge, with a charge time of approximately 1-2 hours.

OperationWhat is the confirmed price?

A stable manufacturer’s suggested retail price could not be verified. The manufacturer’s site shows inconsistent regular-price anchors ($469 and $489) alongside current sale pricing, and CES 2025 debut pricing ($299 base / $319 Pro) differs from both.

OperationHow many faces can it recognize?

The manufacturer states it can remember up to 100 faces.

Availability

Where it's sold

We may earn commission from qualifying purchases. This does not influence classification.

Amazon See current listings

Cite / Embed

Reference the Ropet KAMOMO classification.

Embed the Autonomy Ladder™ mark or copy the citation. The mark is served from this record, so a reassessment updates it on your page automatically.

Get the embed code and citation

124161

268112

Clear space 3222 on all sides

Embed code

Scale proportionally only, up to 2× (536px wide)(224px wide). No recolouring, cropping, rotating, or re-typesetting.

Citation

Robovations. "Ropet KAMOMO: Autonomy Level I (Manual Automation)." Autonomy Ladder™, assessed September 2026. https://robovations.com/robots/ropet-kamomo/

Link it. The mark must point at this record. An unlinked mark is not authorised.
Hot-link it. Never download, self-host, or re-upload the image. A saved copy stops updating, and a mark that has stopped updating is a claim we no longer stand behind.
A classification, not an endorsement. It may not appear beside language implying Robovations recommends, rates, or approves the product.

Where to next

Three ways to keep going.

For evaluators

Compare with a peer.

See the Ropet KAMOMO and Yukai Engineering Qoobo side by side: autonomy level, Robovations Score, readiness, and price. Classification, not a verdict.

See the full comparison

Or compare with Moflin, SwitchBot Bot, ElliQ 3.

For browsers

See every Level I.

Every robot we’ve classified at manual automation. Sorted by readiness, score, and price. Filter by category, manufacturer, or release date.

Browse the database

Updated weekly with new entries and reclassifications.

Ropet KAMOMO

Will the Ropet KAMOMO fit your application?

The full spec sheet.

Living with it.

How the assessment has moved.

What people actually ask.

Other specialty robots we have classified.

Yukai Engineering Qoobo

Moflin

SwitchBot Bot

ElliQ 3

Where it's sold

Reference the Ropet KAMOMO classification.

Three ways to keep going.

Get the next reassessment.

Compare with a peer.

See every Level I.