Your chosen bike frame and wheelset both affect speed in Zwift, so it makes sense for racers to do a little research and choose the best tool for the job. This is especially true for courses that include significant amounts of dirt road, because Zwift’s gravel bikes perform much better than road bikes on dirt!
On top of that, while Zwift’s 4-star rating system for weight and aero is simple and easy to understand, performance can vary even among frames with the same star ratings, because the stars are just an approximation. Plus, Zwift’s addition of bike upgrades means racers need to consider how a frame will perform when fully upgraded, not only when it’s first purchased.
Below, you’ll find charts ranking all the gravel bike frames and wheels based on their performance on flat roads and climbs. We also dig into how they perform in the dirt, including a chart comparing gravel bike performance to top road, TT, and MTB setups.
Table of Contents
Charts last updated April 10, 2026
Gravel Frame Tarmac Performance
We begin with a simple chart showing how much time each gravel frame with the basic Zwift Gravel wheels will cost you compared to the stock Zwift Carbon frame with 32mm Carbon wheels across 1 hour of riding at 300W (4 W/kg) on tarmac.
A few helpful notes:
- These are “stage 0” frames, meaning no performance upgrades have been applied.
- Relative performance on tarmac is the same as relative performance on dirt. That is, if a gravel frame is faster than another gravel frame in these tarmac tests, it will be faster on dirt as well.
- The times below are all negative because all gravel frame+wheel combos are much slower on tarmac than even the most basic road bikes. The faster your gravel frame, the less time you’ll lose.
- Mouse over the bars for specific data. These numbers show how many seconds each particular gravel frame loses compared to the basic road setup. They also let you compare one gravel frame to another. For example, in our flat test, the Cannondale SuperX LAB71 loses 112.4 seconds, while the Lauf True Grit loses 148.7 seconds. That means the Cannondale is 36.3 seconds faster than the Lauf across an hour of flat tarmac.
- The chart is sorted from the best-performing frames to the worst, and you can sort based on flat performance or climb performance.
Sort Chart By:
Takeaways
The most obvious takeaway is that gravel setups are much slower on tarmac than road setups. Even the most aero gravel bike still loses 109.1 seconds to a basic road bike, and the best climbing gravel bike loses 118.7 seconds across an hour of climbing.
Comparing gravel frames to gravel frames, you can see that the latest additions to Zwift’s Drop Shop are the clear top performers in the flat test, taking the top 4 spots. The climb test is a bit mixed, with the super light Specialized Crux in second place and the Canyon Grail essentially tied with the new Grail SLX. But apart from those two frames, the other top 4 spots go to the new gravel frames.
Gravel Frame Performance, Stacked
This chart uses the same data as above, but combines the flat and climb times, sorting by their sum, so frames are ranked based on total time cost across both tests. This chart lets you easily see the best all-around performers, and whether their performance leans to the aero or climbing side.
Takeaways
Nothing too earth-shattering here… four of the top five slots go to the newest bunch of gravel frames, with the Cannondale SuperX LAB71 delivering the best all-around performance.
Gravel Wheels Tarmac Performance
Like the frame chart above, this simple chart shows how much time a particular gravel wheelset (paired with the Zwift Gravel frame) will cost you across 1 hour of riding at 300W (4 W/kg) on tarmac compared to the stock Zwift Carbon road setup.
On Zwift, all gravel wheels are much slower on tarmac than even the most basic road wheels. So the numbers below are all negative, meaning you’re losing time to the stock road setup. The faster your wheels, the less time you’ll lose.
The chart is sorted from the best-performing wheels to the worst, and you can sort based on flat performance or climb performance.
Show Chart For:
Takeaways
On the flats, the two newest gravel wheelsets are tied at the top of the podium, while all the other wheelsets are ~13 seconds slower.
In terms of climb performance, you can see there isn’t much variation from the fastest to slowest wheels. Still, the two newest wheelsets sit atop the chart, while all the older wheelsets turn in essentially the same time, apart from the Zwift Gravel being just a touch slower.
The simple takeaway here is: use the Roval Terra Aero CLX or Zipp 303 XPLR SW wheels.
Jungle Performance – Gravel Frames + Wheels
What about performance in the dirt, where gravel bikes really shine?
Here are two simple charts showing how each gravel frame and wheelset performs in the Jungle. This clearly shows which frames and wheels perform best in the dirt.
Gravel Frames
Time savings vs Zwift Carbon Frame + 32mm Carbon Wheels across 1 hour of Jungle Circuit riding at 300W (4 W/kg). All tests used the Zwift Gravel wheels.
Gravel Wheels
Time savings vs Zwift Carbon Frame + 32mm Carbon Wheels across 1 hour of Jungle Circuit riding at 300W (4 W/kg). All tests used the Zwift Gravel frame.
Takeaways
The first thing to notice is that gravel setups are much faster than road setups on Zwift dirt. More on this below in the “Jungle Top Performer Shootout” chart, but you can broadly estimate that road bikes lose time to gravel bikes in the dirt at twice the rate that gravel bikes lose time to road bikes on tarmac. This is useful information when racing on routes that include both tarmac and dirt surfaces.
The overall ranking of the gravel frames isn’t surprising, with the four newest racing frames taking the top four slots. The same goes for the wheels – while it may look like the two new wheelsets (Roval Terra Aero CLX and Zipp ZIPP 303 XPLR SW) are much faster than the other wheelsets, it’s just a chart scaling trick. Their time gap in the dirt is about the same as on the tarmac.
The simple takeaways here are: buy one of the new gravel frames and new gravel wheelsets if you’re looking to race in the dirt.
Jungle Top Performer Shootout
Our final chart shows Jungle performance for top setups in Gravel, Road, TT, and MTB. People often want to know how one type of setup (eg, TT) performs vs another (eg, Gravel) in dirt, or on a mixed-surface route. There’s no way to quantify this precisely due to race dynamics (drafting) and power fluctuation, but at least the chart below shows how each type of bike broadly performs relative to the other in the Jungle.
Keep in mind the data displayed is seconds saved across an hour of riding, compared to the time of the Zwift Carbon frame paired with the 32mm Carbon wheels.
Use the toggle to see data for stage 0 frames or fully-upgraded stage 5 frames.
Show Chart For:
Takeaways
Gravel setups are the clear winner on dirt courses, beating the next contender (a TT setup) by over 2 minutes across an hour of riding when comparing stage 0 setups. Gravel bikes even come out well ahead when comparing an un-upgraded gravel bike to a fully-upgraded TT rig!
Cervelo Aspero 5 Note
Please note that all speed test data on this page for the Cervelo Aspero 5 reflects an updated (slightly lighter) version of the frame slated for release in Zwift version 1.112 on April 21. Read more about the Aspero 5 here >
Testing Methodology
The data above was generated using our standard bot testing procedures. Our bot rides alone on Zwift, set to 75kg weight, 183cm height. The aero (flat) test route is Tempus Fugit, the flattest course on Zwift, while the weight (climbing) test is done on Alpe du Zwift since it’s a long, steep, and fairly steady climb.
All of the data in these charts (and much more!) can be found in our public Google sheet of test data, and/or our new public Jungle Circuit test data sheet.
Questions or comments?
We’d love to know what you think of these charts, including suggestions for improvements. Share your comments below.
The Pinarello seems to be missing from one of the charts. Oversight?
Turn your phone sideways and it appears
Weird. The data is there but the label isn’t. Still it’s one of the top contenders and in my garage.
It’s the annoying way the charts change layout for mobile. I’ll adjust it to fix…
Making the aspero lighter has made it slower, I guess it was benefitting from high descending speeds with its weight and aero and has lost more there than it gained in the climb
Thank you for this! It is really helpful in deciding how to send those recipes drops. Any chance you will update in the future with info for 2 w/kg rider as you have with the other bikes? Or can you speculate based on what you know from previous tests?
I’ve been riding the New Cannondale Lab71 + Zipp 303 XPLR SW for weeks now, upgrading it, and I had really noticed how much faster these new gravel bikes are in comparison to the older models. Which were real slogs on the road. These results reflect what I’d already noticed. Glad I picked the best overall frame from the start. Thanks Eric for these comprehensive tests!
Does upgrade level change anything between the gravel bikes?
No, they all upgrade equally, so at max upgrade level their charts would basically match what is already above.
The new Diverge is a mid-level upgrade (and the new Cervelo is too, temporarily), so they will upgrade FASTER. But those upgrades boost their speeds exactly the same as the other bikes.
Just want to say that this data is really appreciated, thank you very much!
Happy to help!
In real life, people will often put road wheels and tires on their gravel bikes and get some pretty fast bikes on asphalt as a result. It’s too bad we can’t do the same in Zwift. Maybe if we could, that would encourage people to ride and upgrade gravel bikes more.
Eric, I know you can’t test this directly, but with your understanding of how the game works, do you think it’s the wheels (tires!) that mostly slow the gravel bikes down on asphalt and speed them up on gravel?
And how about the Canyon Inflite, looking like a dog everywhere, despite quietly wining more rainbow jerseys and UCI pro races than any other bike in the modern era!
I have a gravel wheelset for my Domane SL6. It’s a comfy roadie and a decent gravel bike. Saved a ton of money having two bikes for the price of 1.125 bikes! This would make zwift *very* interesting with a lot of permutations and combinations but holy moly Eric might get buried under analysis paralysis and charts!
I ride a 2023 Pinarello Grevil, SRAM 1x, 46-tooth from ring, Eagle derailleur with Force E1 setup, 10-50 cassette. I have two sets of 9Velo wheels, one for road and and for gravel. Bike rides great for both, the bike is definitely better than I am, so the only thing holding me back is me!
So their mistakes with the Aspero will only make it slower after the next patch ?
No – it’s faster (on flats and climbs) after the patch than before. The patch just removed some weight.
But here, before the patch, it was apparantly faster than the Cannondale in the jungle.
https://zwiftinsider.com/cervelo-aspero-5/
Very observant!
We actually modified the Jungle test methodology after that Cervelo post you linked to was posted. The new numbers are accurate. The old ones (used a few posts that will get revised/reposted) weren’t quite accurate.
Do you mean Aspero 5 won’t be the fastest gravel frame in Jungle circuit and SuperX Lab71 will be the fastest one after the 21 April update as shown in this post?
Correct.
Hi Eric,
it’s crazy that the Jungle seems to have his own rules 😳🤣 … because normally the white S-Works SL8 should be the fastest road frame but here it’s the black SL8 with ~5s advantage ???
2nd crazy thing is that the new wheels 6560 in red are already 2s faster than the 6560 in white although the red ones are a bit heavier and not equaled with the white ones which is the plan for next week.
Both things doesn’t make sense imo or do you have an explanation for this ?
Thanks again for testing and sharing.
Then again the Jungle test is done on Zwift Gravel wheels. The overall speed would be higher on the Roval wheels, amplifying the importance of aero.
I wonder, would the more aero Cervelo beat the Cannondale if both were paired with the Roval wheels instead?
It is slower on a constant power effort on a rolling course because it gained time on its rivals on the descents. Most races are won on climbs or sprints so in real racing it will be faster
In Markuri islands you have gravel and climbing. Is the difference in times between the top frames bigger there?
Wish there was more gravel routes, if one would like to fully upgrade the bike it would involve repeating same few routes over and over.
You can equip a gravel bike and ride it on any route. I’ve been upgrading a Crux doing pacer bot rides and Tempus Fugit. Gravel bikes are way slower on road, but with the draft and flatness I think it’s still be faster to upgrade than doing gravel over and over.
Wow I feel even better about my 3rd place finish in PAS stage 1. I was on my Lauf!
I believe the Lauf was one of the fastest gravel frames in one aspect at one point in time long ago!
But never on Zwift, sadly! (https://zwiftinsider.com/lauf-true-grit/)
I must have got my wires crossed because I bought this bike at some point, and I only ever buy bikes if they are the best available for something at the time.
Hi Eric. Thanks a lot for the info.
About the Plasma 6560 wheels and new frames/wheels in that new versión 1.112…
Zwift improved the Plasma to the white ones?
Any new frame could beat the SL8?
S-Works SL8 is still the top frame. Zwift is updating the Princeton 6560 Lava wheels to match the White wheels in performance in next week’s update.
I was expecting more of an improvement for the Aspero 5 at the lighter weight. 🙁
Wonder if we are going to get some Yumezi grit comparisons so decisions can be made for next weeks race series.
Working on it. 😜
https://zwiftinsider.com/yumezi-grit-bike-choice/
First of all, thanks for the tests, they are always super helpful and detailed. I picked up the Áspero because I thought it was the best choice, given it’s the most expensive and has the highest star rating. I really expected that after the weight update it would become the most competitive bike. I don’t quite get why the performance is lacking; even if we’re only talking about a few seconds, it’s a bit of a letdown. I was actually planning on getting the Cannondale at first…
Has the Zwift BMX Bandit been tested recently? I recall it being very fast on dirt at one point in time.
Yup. It’s never been good on dirt. See latest Jungle test results here: https://docs.google.com/spreadsheets/d/14VYeP_xipc5qGd0CRsiDvHLph5QHWBgH9C3tXO-EMa0/edit?gid=0#gid=0
I’m paranoid the PAS races are still broken. Had a fantastic race (r3) and smashed my best 10>30 min efforts. Hoping its the mechanics which was forcing me to push to stay with the roadies on the tarmac combined with a lot of light riders on the climb. Anyone else have trust issues with the gravel situation at the moment.
Yeah been racing the PAs series and seems to me my fully upgraded Aeroroad feels faster than a non upgraded gravel Grail on gravel