Narwal
Narwal Freo Z Ultra
- Autonomy
- III Conditional Autonomy
- Score
- 72/100 Strong
- Readiness
- Ready Now
- Price
- $1,499.99list
Comparison · 2 robots
Narwal Freo Z Ultra · ROIDMI EVA
All robot vacuums. Every measure below is plotted against the full class, so a difference can be read against how much the class itself varies.
- Narwal Freo Z Ultra
- ROIDMI EVA
Roidmi
ROIDMI EVA
- Autonomy
- III Conditional Autonomy
- Score
- 51/100 Limited
- Readiness
- Falls Short
- Price
- $899list
Up to three
01 Classification
What each one does on its own.
The Autonomy Ladder is the same five rungs for every robot on the site. A rung describes the conditions a robot handles without you, not how good it is.
V
Generalized
Any task, any setting
IV
Environmental
Handles new obstacles without help
III
Conditional
Finishes familiar jobs on its own
Narwal Freo Z UltraROIDMI EVA
II
Assisted
Runs preset routines, needs rescue
I
Manual
One action, on command
02 Assessment
Where they actually differ.
The Robovations Score is nine inputs across three pillars, each weighted as shown and each scored so that more is better. A robot can lead on one and trail on another, which is the point of showing them apart. Cost of ownership and data handling are reported on each robot's own page and are never averaged into the score.
Capability
Task completion16% of score
Does it finish the job it is sold to do, unattended
7360
- Narwal Freo Z Ultra73
- ROIDMI EVA60
18median 72 across 354 scored robots100
Coverage and navigation10% of score
Does it work the whole area, and hold that between runs
7075
- Narwal Freo Z Ultra70
- ROIDMI EVA75
15median 70 across 344 scored robots100
Edge-case recovery10% of score
Does it recover from the failures its category presents
7625
- Narwal Freo Z Ultra76
- ROIDMI EVA25
18median 50 across 261 scored robots100
Operating envelope4% of score
How far conditions can vary before performance drops
8065
- Narwal Freo Z Ultra80
- ROIDMI EVA65
25median 70 across 320 scored robots100
Dependability
Field failure rate16% of score
Documented hardware failures per owner-year
7046
- Narwal Freo Z Ultra70
- ROIDMI EVA46
6median 70 across 344 scored robots100
Software stability14% of score
Whether updates fix more than they break
6540
- Narwal Freo Z Ultra65
- ROIDMI EVA40
15median 65 across 275 scored robots100
Support longevity10% of score
Committed parts, service and updates, with dates
630
- Narwal Freo Z Ultra63
- ROIDMI EVA0
0median 63 across 354 scored robots100
Ownership burden
Routine maintenance12% of score
Hands-on time it asks for in normal operation
9150
- Narwal Freo Z Ultra91
- ROIDMI EVA50
22median 68 across 355 scored robots100
Consumables8% of score
Parts that must be bought to keep it working
6175
- Narwal Freo Z Ultra61
- ROIDMI EVA75
30median 69 across 354 scored robots100
03 Measured against the class
How far apart is a long way?
A number only means something next to the spread it sits in. Each track below is the real range across every classified robot in the class, with the median marked.
Price193 with a published figure
$1,499.99$899
- Narwal Freo Z Ultra$1,499.99
- ROIDMI EVA$899
$129.99median $699.99$1,799
Suction (mode not stated)169 with a published figure
12,000 Pa3,200 Pa
- Narwal Freo Z Ultra12,000 Pa
- ROIDMI EVA3,200 Pa
1,000 Pamedian 8,000 Pa36,000 Pa
Runtime168 with a published figure
180 min220 min
- Narwal Freo Z Ultra180 min
- ROIDMI EVA220 min
60 minmedian 180 min360 min
Weight171 with a published figure
9.9 lbs9.3 lbs
- Narwal Freo Z Ultra9.9 lbs
- ROIDMI EVA9.3 lbs
3.3 lbsmedian 8.2 lbs36.4 lbs
Height189 with a published figure
4.3 in4.2 in
- Narwal Freo Z Ultra4.3 in
- ROIDMI EVA4.2 in
2.6 inmedian 3.8 in21.4 in
Not compared, because fewer than two of these robots carry a published figure: Charge time, Noise.
04 Ownership
What the numbers leave out.
The trade-off
What each design gives up to be good at what it is good at.
Narwal Freo Z Ultra
Avoidance down to 5 mm keeps the robot out of trouble and out of the debris: anything the vision system classifies as an object is driven around rather than collected, so small items left on the floor at run time are still there afterward.
The washing station takes away the pad chore and hands back a water chore: pads are washed and dried at the station instead of at the sink, in exchange for refilling the clean tank and emptying the dirty tank on the same rhythm.
ROIDMI EVA
The dock is designed to self-empty and self-wash mop pads, but an independent review found the design under-delivered on both counts in practice, before the manufacturer’s closure removed any path to a fix.
Remaining retail inventory can still be found through some third-party channels, but buying it now means buying a robot with no manufacturer support of any kind.
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.
Narwal Freo Z Ultra
- Homes carpeted mostly in deep shag
- The mops lift 12 mm when the robot detects carpet, which clears low and medium pile. Narwal publishes no maximum pile height and no shag result, so performance on deep shag is unverified rather than documented.
- Households that will not run cameras indoors
- Obstacle recognition depends on two 1600×1200 RGB cameras with a 136 degree field of view, mounted on a robot that drives through every room it cleans. Recognition runs on the robot's dual AI chips rather than in the cloud, but for a household that rules out indoor cameras the sensing stack itself is the disqualifier.
- Budget-conscious buyers seeking entry-level autonomy
- At $1,299-$1,499, this positions above mid-tier LiDAR models. Buyers satisfied with Level III autonomy on a budget should consider more affordable alternatives.
ROIDMI EVA
- Buyers wanting any form of manufacturer support
- Roidmi has ceased operations and discontinued product maintenance, returns, and after-sales service.
- Large homes needing a single uninterrupted cleaning run
- An independent review found the dustbin fills too fast to complete a full-floor clean without manual emptying.
- Buyers relying on the dock to fully dry mop pads unattended
- The same independent review found pads remained damp more than 12 hours after the drying cycle.
What stays your job
The upkeep ownership hands back to you, whatever the autonomy level says.
Narwal Freo Z Ultra
- Weekly: Refill the station’s clean-water tank and empty the dirty-water tank.
- Weekly: Clear hair and carpet fiber from the main brush and side brush.
- Monthly: Clean or replace the primary vacuum filter to maintain suction.
- Every 4-6 months: Replace the main brush when visible wear develops (approximately $18-25 per brush).
- As needed: Replace the mop pads once the pile has matted or they stop coming out of the wash cycle clean.
ROIDMI EVA
- Cadence: Empty the dust bag roughly every 60 days, per manufacturer specification.
- Water tanks: Refill and empty the dock’s dual clean/dirty water tanks.
- Mop pad drying: An independent review found pads remained damp more than 12 hours after the dock’s drying cycle, which in practice may require manual intervention.
- Setup: Define no-go and no-mop zones in the app.
05 Evidence
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 |
|---|---|---|---|
| Narwal Freo Z Ultra | Verified | 9 | Aug 2026 |
| ROIDMI EVA | Discontinued | 3 | Aug 2026 |
Add a robot
Comparing across classes is allowed — the page will switch to per-class positioning and tell you it has.