2026 Ultimate Robot Vacuum and Mop Comparison

In 2025, robotic vacuums and mops saw a ton of new innovations like threshold leaping, bagless base stations, roller mops, and all new navigation systems. But in 2026, we’re mostly just seeing improvement and refinement of those same designs. In this video, I’ve got seven flagship vacuums from Roborock, Dreame, MOVA, eufy, Narwal and Ecovacs and we’re going to figure out if the new models are worth their premium price tags or if buying last year’s model at a steep discount makes more sense. As always, there are no sponsored reviews on this channel.

Carpet Vacuuming

Starting out with carpet vacuuming performance, I mix 10 grams each of flour, salt, flax seed, and rice to simulate different types of dust and debris and I sprinkle that onto my high pile carpet. I then send each vacuum on a 2-pass, vacuum only run on their highest suction power and I weigh the dust bins before and after the run to determine the total amount of debris pickup. Of course, I do a thorough vacuuming in between each run to pick up any leftover debris.

Starting with the least expensive vacuum in this year’s video, the Roborock QREVO Curv 2 Flow is Roborock’s first robotic vacuum with a roller style mop, but for vacuuming it includes their DuoDivide brush roller system along with 20,000 pascals of suction. The QREVO Curv 2 Flow cleaned for 48 minutes and picked up 26.9g, or 67.3% of the flour, salt, flax, and rice mixture while using 47% of its battery. After the auto empty process, the QREVO Curv 2 Flow had 2.3g left in its dust bin including a small amount in the bin itself, but it was mostly flour caked into the air filter.

Next for $1099 is the MOVA Mobius 60, which has MOVA’s TroboWave DuoBrush, 30,000 pascals of suction and a lower gate on the brush housing that is supposed to increase the effectiveness of the suction power. Unfortunately, even with all that tech, the MOVA only managed to pick up 10.8g or 27% of the debris mixture in its 38-minute run where it used 42% of its battery, and needless to say that is not a strong showing. Also concerning was that after the auto empty process there was still 4.6g of debris left in the bin which was heavily caked into the air filter and would have significantly reduced suction power.

After that for $1359 is the Dreame X60 Max Ultra Complete which has a similar brush roller design to the MOVA with the suction enhancing gate on the brush housing, but with a slightly redesigned double brush roller, and 35,000 pascals of suction. The Dreame X60 Max Ultra Complete cleaned for 28 minutes and picked up 26.2g, or 65.5% of the debris, while using just 29% of its battery, which is substantially better than the MOVA and not quite as good as the QREVO. But concerningly, after the auto empty process, the Dreame X60 still had 7.1g of dust in its bin which was a mix of flax and flour caked into the filter and loose debris in the bin.

Next for $1389 is the Roborock Saros 20, which is the 2026 version of the Roborock Saros 10R, and while there were some slight changes to the DuoDivide brush roller system, the biggest upgrade was in suction, going from 22,000 pascals on the 10R to 36,000 pascals on the Saros 20. But in my years of testing vacuums I’ve found that carpet pickup is mostly about brush design rather than suction, and the Saros 20 cleaned for 24 minutes, which was the fastest by far, and only used 21% of its battery, while cleaning 25.1g or 62.8% of the flour, salt, flax, and rice mixture, which is actually about 3g less than last year’s Saros 10R despite the increase in suction, putting the Saros 20 in 3^rd place so far. After the auto empty process, the Saros had 1.9g left in its bin, which again, was mostly flour caked into the air filter.

After that for a price of $1499 is the just released Narwal Flow 2, which as the name implies is the successor to last year’s Narwal Flow, the highest performing mop that I’ve ever tested but struggled with carpet vacuuming performance. This year the Narwal Flow 2 cleaned for 25 minutes and used 24% of its battery to pick up 18.6g of debris, which is 46.5% of the 40 grams possible, and while that is 6% higher than last year’s Narwal Flow, it still puts the Flow 2 in 2^nd to last place so far. However, after the auto empty process, the Narwal Flow 2 had just 0.1g left in its bin which is extremely impressive, and the air filter was almost completely clear. That’s important because it will help prevent the Narwal Flow 2 from losing suction over time.

After that we’ve got the $1499 Ecovacs X12 OMNICyclone. The X12’s single brush roller remains mostly unchanged from last year’s X11, but the X12 did get a slight boost in suction from 19,500 pascals up to 22,000 pascals. In this year’s test the X12 OMNICyclone cleaned for 46 minutes and used 59% of its battery to pick up 17.9g or 44.8% of the debris mixture, which does put it slightly behind the Narwal Flow 2, but it is worth noting that it cleaned for almost twice as long and used over twice as much battery power. However, even though it had lower than average pickup, it had a much higher than average auto empty efficiency, leaving just 0.8g in its dustbin after the auto empty process. It did have some flour caked in the filter, but much less than the rest of the vacuums other than the Narwal.

Last, the most expensive vacuum in this video is the new eufy OMNI S2 which shares the same general shape as the eufy S1 Pro that was my top pick in 2024, but the S2 is quite a bit bigger and has a completely redesigned divided brush roller, a dust bin that resembles a stick vac, and 30,000 pascals of suction. In the carpet pickup test the eufy S2 cleaned for 34 minutes, used 52% of its battery, and picked up a total of 31.5g or 78.8% of the debris mixture, which not only catapults it into first place in this video, but is also the highest score I’ve recorded since switching to my current testing equipment three years ago. After the auto empty process, the eufy had 1.5g of debris left in its bin, but its main air filter was completely clean thanks to its unique dustbin design.

Comparing this year’s results to all the other vacuums I’ve tested, the eufy S2 sits on top but the rest of the vacuums from this video fall significantly below the more budget friendly MOVA P50 and Dreame L40S Ultra CE that previously held the top two spots for carpet vacuuming performance.

Hard Floor Vacuuming

Moving on to hard floor vacuuming, I used that same 40-gram mixture of flour, salt, flax, and rice. Historically, vacuums score really well on this test, between 95 and 100% pickup, but something I’ve never really reported on before is the feel of the floors on my bare feet after a cleaning, because sometimes even vacuums with high pickup percentages leave the floors feeling gritty. So, in addition to measuring their actual pickup performance, which you can see here with all the vacuums scoring between 96 and 99.5%, I also gave each vacuum a more qualitative score for how the floors felt after the cleaning. Some notable results were the Roborock QREVO Curv 2 Flow that had a middle of the road pickup score due to the fact that it knocked some rice grains under my refrigerator, but still had a very clean floor feeling, and on the other end of the spectrum, the MOVA Mobius 60 picked up 99.3% of the debris mixture, but left a thin film of flour dust on the floor which made the floors feel pretty dirty.

If we combine the carpet and hard floor scores together you can see that the eufy S2 finishes in first with an 88.6% combined score, which puts it just barely into 1^st place for all the vacuums I’ve ever tested, but it was held back by a slightly lower than average hard floor pickup score caused by its side facing exhaust that pushes debris out of the way as it passes from left to right.

But just looking at debris pickup performance doesn’t tell the whole story, because one of the big reasons why less expensive vacuums have better carpet pickup performance is, unlike flagships, they aren’t as focused on avoiding hair tangles and therefore have completely different brush roller designs.

Hair Pickup and Tangles

So next we’ll look at hair pickup and tangle avoidance. In my initial test I count out just 30 strands of 12” human hair and spread them on my bathroom floor before sending each vacuum on a single pass vacuum only run, and then I carefully inspect the floor and the vacuums to find any leftover or tangled hair. In this test, all the robots completely cleaned the hair off the floor, and only the MOVA Mobius 60 and Dreame X60 had a single hair on the vacuums themselves.

My second test is more of a stress test where I spread 1 gram of 16” synthetic blue hair and two cotton balls separated into tufts to simulate pet fluff, and after the vacuuming run, I collect any hair that’s left on the carpet and inspect the robots themselves for tangles.

The best performer by far here was the Narwal Flow 2, which cleaned 100% of the hair and left just a single tuft of fluff on the carpet. It also managed to stay completely tangle free, with just a single tuft of cotton stuck in the wheel, which is great to see since the original Narwal Flow was the first vacuum I’ve ever tested to score perfectly on this test and it’s nice that the Flow 2 maintains that same level of performance.

I originally thought the eufy OMNI S2 got an actual perfect score with nothing left over on the carpet or on the robot, but after reviewing the footage you can see that side facing exhaust acting like a leaf blower, and it actually just pushed all the fluff under the couch instead of picking it up. But it did do a good job picking up the blue hair, and it also ended up completely tangle-free on the vacuum itself which is still a good result.

The MOVA Mobius also performed decently in terms of hair pickup, leaving just a few broken blue hairs and a small bundle of fluff on the carpet, but it had a large, tangled ball of hair stuck in its wheel well.

The Roborock Saros 20, QREVO Curv 2 Flow, and Dreame X60 Max Ultra Complete had pretty average results in terms of hair pickup and tangling, and the Ecovacs X12 OMNICyclone really struggled and not only left a significant amount of hair and fluff on the carpet, but it also failed to get any of the hair into its dustbin and left a matted mess clogged behind the brush roller.

Corner Vacuuming

I also tested corner vacuuming ability, by placing 1 teaspoon of dried breadcrumbs in four different hard to reach locations and awarding 1 point for a partial clean and 2 points for a completely cleaned pile. The eufy OMNI S2 and Roborock Saros 20 scored 6 out of 8 possible points by completely cleaning three of the four piles but failing to clean under the refrigerator. The Dreame X60 Max Ultra Complete also scored 6 points by completely cleaning two piles and partially cleaning both the inside cabinet corner, and under the refrigerator.

The QREVO Curv 2 Flow was too tall to fit under the counter stool, resulting in a score of 4 out of 8, the Ecovacs completely skipped the pile by the dishwasher and scattered the pile on the inside corner for a score of 3 out of 8, and the MOVA decided it couldn’t fit under the stool even though it has a retractable LIDAR turret and it also completely missed the inside corner pile and the refrigerator, for a final score of 2 out of 8.

Combined Vacuuming Scoring

If we combine all of the vacuuming scores together, that puts the eufy OMNI S2 on top, only held back by the singular issue where the powerful suction motor exhausts out of the side of the vacuum instead of the back which gives it a leaf blower effect with extra light debris. But the OMNI S2 recorded the highest ever carpet vacuuming score, had excellent hair tangle avoidance, and a great corner cleaning algorithm.

The surprising second place vacuum was the lower priced Roborock QREVO Curve 2 Flow which had above average carpet and hard floor pickup, good tangle avoidance, and above average corner cleaning ability, but was too tall to fit under the stool which knocked down its overall corner pickup score.

Then third place was a 3-way tie between the Dreame X60 Max Ultra Complete, Narwal Flow 2, and Roborock Saros 20, which were mostly just average when it comes to vacuuming performance.

Mopping Performance

Next, it’s time to look at mopping performance. For my first test, I spread 10mL of Florida mud on my LVP flooring and dry it with a hair dryer. I then send the robots on a 2-pass combined vacuum and mopping run with the mopping water level set to maximum. After the run, I spray the stain with water and wipe it clean with a paper towel, and then I set the towel aside to dry.

In this video, there are two different mop types. The Roborock Saros 20, MOVA Mobius 60, and Dreame X60 Max Ultra Complete all use dual spinning mopping pads that periodically clean themselves at the base station, while the Ecovacs X12 OMNICyclone, Narwal Flow 2, Roborock QREVO Curv 2 Flow, and eufy OMNI S2 all use roller style mopping pads that are constantly being cleaned by wetting the roller with clean water and then scraping the dirty water out into an onboard dirty water storage tank. 

The MOVA Mobius 60 also has another trick where it actually has three different sets of pads that it can interchange for different flooring types, but in my testing it seemed like there was only a very minor difference in cleaning ability between the three pad types, and I would have liked to see a more abrasive option included for more durable flooring types.

The Ecovacs X12 OMNICyclone also has a unique feature where it can spot stains on the floor and spray them with front-facing water jets to let them soak before mopping, but in my mopping test the X12 did not identify the large mud stain as an area to pre-wet with its jet spray.

I’m not sure if this batch of mud was harder to clean than previous years, because we did just have our annual oak pollen drop, but it seemed like the mops struggled more than normal. Looking at the final cleaned stains, the Dreame X60 Ultra Complete had the hardest time and really only cleaned the stain on the seams of the LVP. The MOVA Mobius 60 was next and also struggled to clean the middle of each plank. The Ecovacs X12 OMNICyclone performed surprisingly poorly for a roller mop, the Roborock Saros 20 was slightly better than that, the eufy OMNI S2 did an overall decent job but still left noticeable mud on the floor, the Narwal Flow 2 did very well and the floor was visibly clean, but the Roborock QREVO Curv 2 Flow did by far the best job and left a floor that was not only visibly clean, but also passed with a completely clean paper towel test.

Speaking of the paper towel test, here’s what those looked like for this round of mops.

Out of all the robots in this video, I would say I’m most disappointed in the Ecovacs X12 OMNICyclone that didn’t activate its jet spray system at all on this run. Just out of curiosity, I re-ran the test in its forced jet spray mode and the results were excellent, finishing somewhere in between the Narwal Flow 2 and QREVO Curv 2 Flow for both the visual score and paper towel test, but Ecovacs really needs to get jet spray active and working in normal mopping modes.

Corner Mopping

I also tested corner mopping ability by putting a smear of hot sauce in five separate, hard to reach locations in my kitchen and like the corner vacuuming test, I give each vacuum 1 point for a partial clean and 2 points for fully cleaning each spot. But importantly, I award no points for only cleaning the stain with the side brush or roller, only for cleaning with the mops.

The surprising worst performer on this test scoring just 2 out of 10 was the eufy OMNI S2, which completely missed every single corner stain and bumped into and moved the stool while cleaning under it, which is really surprising since the eufy S1 Pro was one of the most confident navigators I’ve ever tested. It seems like the S2 doesn’t understand its own dimensions which makes it clumsy and affects its ability to properly hug corners using its extendable mop.

Narwal Flow 2 also had issues with clumsy navigation, and despite having a new extendable mop, it still struggled to adequately clean corners, scoring 5 out of 10 possible points.

The Ecovacs X12 OMNICyclone ran its side brush through every single stain, but only partially cleaned most of the stains with its mop, scoring 6 out of 10.

Roborock Saros 20 could have scored significantly better if it used its extending mopping pad more when going around corners but ended up with a score of 6 out of 10.

The MOVA Mobius 60 also had a bunch of partial scores, earning 6 out of 10 total, but on a positive note, it did decide to clean under the counter stools this time, unlike the corner vacuuming test when it determined that it couldn’t fit under them.

The top score went to the Dreame X60 Max Ultra Complete that liberally used its extending mopping pad and was slow and methodical around corners, fully cleaning two of the stains and partially cleaning the other three for a score of 7 out of 10.

Mop Lifting and Covering

The last thing to test with mopping is whether the robots can perform a combined vacuum and mopping run without getting carpets wet in a room with mixed flooring.

Out of all the robotic mops in this video, only the Roborock QREVO Curv 2 Flow was able to completely avoid getting the carpets wet by using a retractable plastic cover that goes over the roller mop.

None of the robots with spinning mopping pads were able to lift them high enough to avoid dragging them on the carpet, but all three are able to drop their mop pads at the base station and prioritize carpet cleaning first, so the only time mop lifting would matter is if the vacuum had to pass over a carpeted area to get from one hard floor to another.

Neither the Narwal Flow 2 nor the eufy S2 could lift their roller mop enough to avoid getting the carpets wet, and neither has a plastic cover for their mop roller like the QREVO Curv 2 Flow. While the Ecovacs X12 OMNICyclone has a retractable cover for its roller mop, it was still getting the carpets really wet, and I was able to figure out the cause. The back of the robot gets soaked during mopping, which then gets transferred to the carpet.

Combined Mopping Scores

So then if we average the three mopping categories, the Roborock QREVO Curv 2 Flow comes out on top with by far the best actual mopping score, the best carpet crossing score, and a middle of the road corner mopping score. While the Narwal Flow 2 and eufy S2 did a relatively good job with stain cleaning, their poor corner cleaning and inability to cross carpets negatively affected their overall scores.

Object Avoidance

Moving on to object avoidance, I set up a room with a pair of untied shoes, a loose extension cord, and a pile of plastic dog poop. I award the robots 1 point for touching either the shoes or extension cord without getting stuck, 2 points for avoiding them completely, and an all-or-nothing 2 points for avoiding the dog poop.

The top performer, with a perfect 6/6 score was the Ecovacs X12 OMNICyclone, which is a major victory for Ecovacs which has traditionally struggled on this test. The MOVA Mobius 60 almost scored a perfect score, avoiding every object for nearly 45 minutes before accidentally getting too close to the dog poop at the end of its run, ending with a 4 out of 6. The Dreame X60 Max Ultra Complete ran directly over the dog poop at the beginning of the run and then successfully avoided it for the remainder of its cleaning task for a score of 4 out of 6. The Roborock Saros 20, eufy S2, and Narwal Flow 2 all ran over the dog poop and gently nudged the shoes for a score of 3 out of 6, while the Roborock QREVO Curv 2 Flow showed why it’s considered a less flagship level vacuum and had a lot of trouble in this test, bumping into the shoes, completely ignoring the loose cord, and running over the dog poop multiple times for a score of 2 out of 6.

Threshold Crossing

Another thing that is becoming more standard on flagship vacuums is their ability to cross very tall thresholds. I tested three different heights of thresholds, each with a ramped side and a flat side. The Roborock Saros 20, Mova Mobius 60, and Dreame X60 Max Ultra Complete were all able to traverse the 50mm threshold in both directions. The Narwal Flow 2 and eufy S2 were able to traverse the 30mm threshold in both directions. Both the Ecovacs X12 OMNICyclone and Roborock QREVO Curv 2 Flow were able to cross the ramped side of the 30mm threshold but failed to cross the flat side. All the vacuums easily crossed the more reasonably sized 13mm threshold.

Vacuum Noise Level

The last thing to look at on the robots themselves before we move onto apps and smart home platforms is their noise level in case you want to have these running while you’re in the house. For this test, I set each vacuum to their second highest suction power and here they are from the loudest to the quietest.

Apps and Smart Home

The final thing to cover is the apps and smart home integrations for these robots. For a long time, the apps were mostly the same, but as features have been gradually added, the Roborock app has risen to the top as being both user friendly and feature rich.

The recent changes to the MOVA and Dreame apps have made the maps less detailed in favor of a more square-based design which makes it harder to set up no go zones and recognize features in rooms. As they’ve added more and more features, they have become buried in sub-menus that make them more difficult to understand, find, and use.

Similarly, the Ecovacs app has lots of options buried deep in menus and has put an emphasis on their AI Agent Mode and Yiko voice assistant, which is fine, but if I’m accessing the app, it’s probably because I want to change a setting, not talk to AI.

The Narwal app is fine and actually has a pretty manageable number of options, but I don’t like that the default option for each of the cleaning tasks is to use their automatic “Freo Mode” since in my experience those modes can take an extremely long time to complete if the robot decides that it needs to vacuum or mop a room several times. My number one gripe with the Narwal app is that by default carpets are set to “cross only” which will prevent the Narwal from cleaning your carpets, even on vacuum only mode.

Last, the eufy app is as simple as it can possibly be. The 3D house map works well and is easy to edit, and setting up tasks is extremely simple, but it’s the one app that doesn’t have enough options. As far as I can tell, it doesn’t even have an option to vacuum carpets first before getting the mop wet, which is something that most other brands have been able to do for years now. Out of these seven vacuums, the eufy is also the only one that doesn’t allow you to use the front-facing camera on the robot to drive it around your house remotely, which again is a feature most brands have had for years. Eufy is also the only one that doesn’t support room specific commands when asking it to clean via Amazon Echo, and it also doesn’t have any built-in voice assistant, which I actually don’t mind but it seems like something that the most expensive vacuum in the video should have.

Local Control

However, none of this really matters, because what matters is Matter, and it’s here. For the first time basically all of these vacuums are controllable locally when connected via Matter over Wi-Fi. For my testing, I connected the vacuums to Home Assistant via Matter and was disappointed that there weren’t room specific controls. It turns out that it was possible, but it involved first mapping the vacuum’s segments to Home Assistant areas and then calling the “clean area with vacuum cleaner” action, which seemed a bit cumbersome. So, I went searching for a better way, and after I shared the vacuums with Apple Home, I could select vacuuming, mopping, or vacuum and mop, and I could select any number of rooms.

I also blocked the vacuums from the internet using my router and they remained completely controllable via HomeKit. Whether they can remain disconnected indefinitely is still up in the air, but as of writing this script they’ve been blocked from the internet for 24 hours and are still working. That’s really exciting because this is the first time we’ve had pure local control with high usability for flagship robotic vacuums.

Conclusions and Total Scores

So now it’s conclusion time, and in this chart, I’ve averaged all the individual grades for each category with carpet vacuuming and hard floor mopping given double weight. You can see that the Roborock QREVO Curv 2 Flow that had the 2^nd highest carpet vacuuming score and the best mopping performance ended up in first place. The roller mop cover was 100% effective at keeping carpets dry, making it an excellent option for homes with mixed hard flooring and high pile carpets and it’s a great pick and an excellent value considering it’s priced much lower than the rest of the vacuums in this video. But you should be aware that it’s missing common flagship features like advanced object avoidance, high threshold crossing, and it’s much taller than the Roborock Saros 20, eufy S2, and Dreame X60 Max Ultra Complete, so it’s not as good for cleaning under low furniture.

To me, the eufy OMNI S2 seems like the most flagship level vacuum in this video, recording the highest carpet vacuuming score of any vacuum I’ve ever tested with well above average mopping performance, great hair tangle avoidance, good object avoidance, and Matter connectivity. Its two biggest shortcomings are the fact that it has a side exit exhaust that pushes light debris out of the way and that its corner mopping algorithm makes it look like the OMNI S2 has no idea what its own dimensions are. But both of those issues are fixed with better path planning that can happen via firmware updates. While they’re at it, eufy should also add an option to vacuum all the carpets first before getting the mop wet since it lacks a roller mop cover like the less expensive Roborock QREVO Curv 2 Flow.

If you need a threshold climber, both the Roborock Saros 20 and Dreame X60 Max Ultra Complete easily cleared even the tallest 50mm thresholds. But while neither vacuum had any major shortcomings, they mostly just performed average or slightly above average in all of my tests, which is a bit disappointing for newly released flagships with very premium price tags. If you have tall thresholds, it probably makes more sense to look at last year’s Dreame X50 Ultra that can also clear 50mm thresholds and had very similar vacuuming and mopping performance for $500 less.

As always there are no sponsored reviews on this channel, but I do have links to all the vacuums in this video below and as always, I appreciate it when you use those links since as an Amazon affiliate, I do earn a small commission on the sale at no cost to you.

I’d also like to thank all my awesome patrons over at Patreon for their continued support of my channel, and if you’re interested in supporting my channel please check out the links below. If you enjoyed this video, please consider subscribing to my YouTube channel and as always, thanks for watching The Hook Up.