
Published Jan 11, 2021 · Updated May 6, 2026
Use Zwift custom workouts to grow your FTP with a data-driven, personalized Zwift custom training plan from AI Endurance.
Edit, May 2026: You no longer need any of the manual steps below. Just connect Zwift on the AI Endurance Apps page and your scheduled cycling workout for the day will automatically appear in the Zwift, ready to launch on your trainer. Anything you change in AI Endurance (swapping a workout, moving it to a different day, skipping it) is reflected in Zwift within seconds. The instructions below describing the manual .zwo download and the Documents\Zwift\Workouts\ folder are kept for historical reference only.
You can now execute your AI Endurance workouts as Zwift custom workouts by following these simple steps:
In case you are updating your training plan make sure to delete all AI Endurance .zwo files from your old plan in the workout folder prior to uploading your new workouts. Don't delete the file workouts.files as it is Zwift's way of keeping track which workouts were deleted.
For more information on how to use custom workouts in Zwift, see What's on Zwift.

That's it, now you can execute your AI Endurance rides as a Zwift custom workouts training plan.
See also our Zwift running workouts.
You can also get a taste of some our workouts under
AI Endurance is based on the observation that an optimal training routine can be very different for each individual. That's why one-size-fits all training plans often don't yield the expected results.
With AI Endurance you get truly personalized, data-driven training based on your accumulated historic power data. Our machine learning algorithm is like a 'digital twin' that represents how you respond to different training routines. This allows us to
For your optimized, personalized training plan we take into account
You can always make adjustments to your training plan when real life gets in the way.
Sign up today and get your own personalized training plan to reach your goal FTP!

DFA alpha 1 lets you estimate your aerobic and anaerobic thresholds from heart rate variability, with no lab test and no dedicated test workout. A chest-strap HRV monitor like the Polar H10 is all you need for heart-rate thresholds; add power or pace data and you get corresponding thresholds in watts or pace units too. Here is how it works and how accurate it is.

Before every workout you should know if you're actually ready for it. Everyone responds differently to stress, bad sleep and exercise fatigue - our new recovery model makes data driven decisions about when you should train and when you shouldn't - based on heart rate variability (HRV).

by Markus Rummel. We present the first results of AI Endurance's new capability to calculate Respiration Frequency (RF) from in-activity heart rate variability (HRV) data. RF demonstrates its potential in assessing the validity of HRV threshold determination.

by Stefano Andriolo. Building on previous work, we refine a method to accurately determine the relationship between DFA alpha 1 and power. This method can be used to track fitness and thresholds of an athlete. We find in some cases ramp detection tends to overestimate thresholds, a finding mirrored in recent physiological papers. On the other hand, thresholds based on clustering of DFA alpha 1 values tend to agree well with this new method. We propose a hybrid lab and everyday workout experiment to further study the relationship.