One score ranks you. The other only knows you.

Gravl and ftns.ai both give you a number that goes up. That is where the similarity stops, and the difference between the two numbers is more interesting than it sounds.

Is ftns.ai a Gravl alternative?

Only if what you want is a reason to train, rather than the training itself.

Gravl, built by GAINS AI, does one job carefully. It estimates your one-rep max from your logged history, builds the next session from it, and steps the weight up when the rep ladder fills. It accounts for the equipment in your actual gym, down to which plates are on the rack. It has a library of 300+ trainer-led videos, syncs with Apple Health, Health Connect, Garmin and Strava, and runs standalone on the wrist.

ftns.ai has a coach that will build you a session too — from your goal, your pace, your equipment and the exercises you have ruled out. What it does not have is any of the load science underneath Gravl: no one-rep-max estimate, no plate maths, no progression model deciding that today is the day you add five kilos. If that machinery is what you are shopping for, Gravl is the better answer and this post is not going to argue otherwise.

What ftns.ai does instead is take everything you did — workouts, yes, but also sleep, steps, water, meals, weight, mood, alcohol and caffeine — and turn all of it into a full-colour 3D character built to your own proportions. Your twin changes as you do.

The Strength Score and the Life Score are opposite ideas

One compares you to other people. The other compares you to yourself. That is the entire philosophical gap between these apps.

Gravl calculates a Strength Score from eight sub-scores and benchmarks it against data from people of the same age, sex, body weight and training experience. It is a well-built metric, and a lot of lifters find that context genuinely useful. Knowing your bench is average for your bodyweight is real information.

The ftns.ai Life Score does not do that. It is one number out of a hundred, built from six categories, with sub-bars for each — and it is only ever compared to your own history. There is no rank, no percentile, no cohort, no top one per cent. Nobody else's number appears anywhere in the app, because there is nowhere for it to appear.

This is a real trade-off and we should say so. Comparative benchmarking is motivating for a lot of people. Knowing where you sit relative to your peers gives progress a shape that a personal line chart does not have. If that is what drives you, an app with no comparison in it is worse for you than one with it.

We built the other one for the people who quietly deleted a fitness app the week their number became something that could be ranked. The long version of that argument is here.

Does Gravl already project your progress?

It projects numbers. We project a character. Those are not the same thing, and we would rather be clear than clever about it.

Published walkthroughs of Gravl's onboarding describe a strength projection screen that estimates gains in key lifts over the following months, turning an abstract promise into a concrete figure before you have trained once. It is a smart piece of product design, and we are not going to pretend we invented forward projection.

Future Self is a different object. It renders your twin across six horizons and three timelines — Current Pace, On Plan and Coasting — as a character you can look at, rather than a kilo figure you have to imagine. It is free, unlimited, and it never costs gems.

It is a game feature, not a forecast. It does not predict anything about your health, it guarantees nothing, and it does not tell you what will happen to your body. It shows you a version of your character built from the way you are currently playing. How the six horizons work.

Side by side

Gravl details per its App Store and Google Play listings, gravl.ai and independent 2026 reviews, checked 24 August 2026. Published Gravl pricing differs between sources and regions, so it is not quoted here — check your own checkout.

What ftns.ai is not

Stated plainly, because the category invites the assumption:

• Not a training-optimisation engine. The coach plans sessions. It does not estimate your one-rep max, do plate maths, model fatigue or periodise a block. Apps built for that do it better, and this post says so out loud.
• Not a health intelligence. The .ai is a domain. Your coach is a game feature, and it makes no medical claims and promises no results.
• Not a medical or assessment product. ftns.ai is a game in the Health & Fitness category. It doesn't diagnose anything, doesn't give clinical advice, and doesn't tell you what any of your numbers mean about your health.
• Not social. No social feed, no friends list, no leaderboards, no follower count. Sharing means you export a score card yourself, to wherever you like, when you feel like it. The app never posts on your behalf.

Why your twin updates weekly

Your twin's body recomputes once a week, on the trend, rather than reacting to every session. Mood moves faster, on a separate axis, over the last few days.

That is deliberate. If a single skipped workout visibly changed your character, the model would be lying to you and you would stop believing it by week two. Real bodies move on trends. A character that moves faster than a body is a slot machine, not a mirror.

So which one

• Use Gravl if you lift, you want the app to manage the next set, and knowing how your strength stacks up against people like you is motivating rather than discouraging.
• Use ftns.ai if the barrier is showing up rather than knowing what to do, you want sleep and steps and water to count as much as your lifts, and you want your progress ranked against absolutely nobody.
• Use both if you want a managed programme for the hour and a character for the week. They do not overlap enough to conflict.