The Riven Journal
Product

The Best Automatic Rep Counter Apps for Apple Watch in 2026 (Honest Review)

Honest 2026 review of automatic rep counter apps for Apple Watch — Motra, Gymatic, Rep Up, Fitnexx vs Riven. Real pros, cons, and accuracy.

The Best Automatic Rep Counter Apps for Apple Watch in 2026 (Honest Review)Riven · Product

The best automatic rep counter app for Apple Watch in 2026 depends on what you lift. For free-weight barbell and dumbbell work, Motra (formerly Train Fitness) and Gymatic count reps from wrist motion alone. For a no-look haptic nudge, Rep Up taps your wrist every few reps. And if you want the count plus a read on whether the set was actually hard, that's where Riven separates from the pack. One thing first: the Apple Watch does not count reps on its own. Every option here is a third-party app.

Let me save you the marketing fog. I've lifted for years and I track everything, and I've watched the auto-counting category make real progress while still tripping on the same physics. Here's the honest version.

What is the best automatic rep counter app for Apple Watch?

There's no single winner — there's a best for your exercise mix. Motra leads on breadth (it claims 470+ exercises) but independent testing rated it 3/5 and called it "not yet reliable enough to replace intentional logging" (findyouredge.app, 2026). Gymatic shines on exercise auto-detection and rich per-rep data. Rep Up wins on simplicity. Riven is the only one that counts reps, names the exercise, and tells you if you hit failure.

Here's the catch nobody in the App Store screenshots: Apple's built-in "Functional Strength Training" and "Traditional Strength Training" modes track duration, heart rate, and calories — and nothing else. No reps. No sets. No exercise type. If you want a rep count on an Apple Watch, you install something. Period.

How automatic rep counting actually works

Automatic rep counting works by reading the watch's motion sensors — the accelerometer and gyroscope (the IMU), plus rotation and gravity via Core Motion — and finding the repeating rhythm of your reps. Each rep produces one motion peak. The app detects those peaks and counts them. Some apps add heart rate for context. None of them measure load, barbell velocity, or joint range of motion directly — those get inferred from wrist movement, and inference is where things wobble.

That last point is the whole game. A wrist IMU is genuinely good at velocity, acceleration, and the broad shape of a movement. It's blind to how heavy the bar is and how far your elbow actually traveled. So a curl with a big, clean wrist arc reads beautifully. A leg press where your hands sit still on the handles? The watch sees almost nothing.

This isn't a bug waiting for a patch. It's a law of physics for wrist sensors. Garmin's own native rep counter "excludes many machine exercises entirely" because counting requires wrist movement (the5krunner, Apr 2026). The velocity-based-training literature backs the broad point: a systematic review found 7 of 8 IMU models valid for barbell velocity, but wrist-worn IMUs are generally noisier than dedicated linear position transducers (PMC8038306). Usable signal, noisy regime. That's exactly where all these apps live.

Motra, Gymatic, Rep Up, Fitnexx — the honest rundown

Four apps, three different philosophies. Some try to identify your exercise and count it; one deliberately refuses to identify anything.

Motra (formerly Train Fitness)

The category's flagship. Motra claims 470+ detectable exercises from the watch accelerometer alone, using what it markets as "Neural Kinetic Profiling" built on 8+ years of Harvard / University of Toronto research (Motra). It logs each rep as you move and lets you confirm or correct after the set.

Pros: Biggest exercise library on the list, covers bodyweight through cable, polished app, large user base. Cons: Independent testing rated it 3/5 — "a promising concept that is not yet reliable enough to replace intentional workout logging." Cable exercises, complex movements, and unusual wrist positions confuse it (findyouredge.app). You'll still correct sets.

Gymatic (VimoFit)

Gymatic positions itself as the first app to automatically identify exercises, count reps, and log workouts (Product Hunt). The auto-detection is the part people actually rave about, and the data is deep: rep speed, time between reps and sets, total weight moved, work-versus-rest time.

Pros: Strong exercise auto-ID, genuinely rich analytics, you can teach it movements. Cons: It ends sets when you pause to breathe and undercounts — reviewers report 5 squats logged as 3 (App Store). Pause too long mid-set and it thinks you're done.

Rep Up

The contrarian. Rep Up doesn't try to name your exercise at all. It detects the repeating motion pattern and taps your wrist every N reps — say, every 5 — so you never look down (App Store). Less data to display, less data to get wrong.

Pros: Dead-simple, no screen-staring, works for anything with prominent hand movement (squats, deadlifts, rows, presses) and some stationary-hand moves. Runs on Apple Watch Series 1 and up. Cons: It admits it sometimes over- or under-counts, and you get a haptic count, not a logged, named, analyzable set.

Fitnexx

Fitnexx auto-counts reps from watch motion and runs independently on the watch — no phone tethered. It targets bodyweight work (squats, push-ups, jump rope) plus gym lifts like deadlift and bench, and it exposes a sensitivity slider because counting can drift.

Pros: On-watch independence, sensitivity tuning, solid 4.6/5 from ~30 reviews (App Store). Cons: Small free exercise library, and accuracy can degrade — one reviewer logged "2 of 10" jump-rope reps after an iOS update.

Notice the pattern. Every wrist-motion counter shares the same weak spot — machines, cables, and any movement where the wrist barely moves or sits at an odd angle. That's not four separate flaws. It's one physical constraint showing up four times.

The thing rep counters miss: was the set actually hard?

A rep count tells you how many. It never tells you how hard. And how hard is the number that actually drives results. Counting 10 reps means nothing if you stopped with 6 in the tank. The research is clear here: muscle growth increases as you train closer to failure — though the relationship is non-linear and absolute failure isn't required (Refalo et al., 2023).

So proximity to failure matters. The problem? Lifters are bad at judging it. You think you had two reps left; you actually had five. That gap is exactly what a bare tally can't close.

This is where Riven does something the others don't. It counts your reps automatically from the watch's motion sensors, recognizes the exercise from your wrist motion, and reads your proximity to muscle failure from rep-speed decay — velocity loss across the set, plus heart-rate behavior. When your reps slow down despite the same effort, the muscle is fatiguing, and that's a scientifically grounded fatigue signal. Motra, Gymatic, Rep Up, Fitnexx — none of them output that. They stop at the count.

Worth flagging, because honest reviews shouldn't parrot gym lore: "always train to failure" is oversimplified, and so is "always think about the muscle." A meta-analysis on attentional focus found an external focus gives a measurable acute strength benefit (SMD 0.34, more for lower body), while long-term differences between internal and external focus wash out (Grgic et al., 2021). Knowing how close you are to failure is more useful than a binary "go to failure" rule.

On the counting itself, here's Riven's honest take. In our own testing, the rep count lands within about one rep on the large majority of sets. Not perfect — no wrist counter is. When it's off, you fix it with a twist of the Digital Crown on the set-review screen. And the same physics that limits everyone limits Riven: fixed-wrist machine and cable holds are the hard cases and may need manual entry. One more bit of honesty — you still tap once to start and end a set. The reps and the exercise ID are automatic; the set boundaries aren't yet.

Which one should you pick?

Pick by use-case, not by hype:

  • Free-weight lifter (barbell/dumbbell) who wants count + effort: Riven. You get auto rep counting, exercise ID, and a failure read in one tap. riven.fit
  • Want the biggest exercise library and don't mind correcting: Motra.
  • Love deep per-rep analytics and exercise auto-detection: Gymatic — just don't pause too long mid-set.
  • Hate looking at your watch, want a haptic nudge: Rep Up.
  • Bodyweight focus, want on-watch independence: Fitnexx.

A note on platforms outside Apple. Garmin counts reps natively but does it unreliably — the5krunner reports miscounts on low-wrist-movement exercises and manual correction mid-set (the5krunner). Whoop's 2026 Passive MSK estimates strength load for Strain from wrist kinematics and body mass — it does not count reps at all (the5krunner, Feb 2026). Fitbit, Samsung, and Oura have no notable consumer rep counter in 2026.

One more reality check on adherence. If you're inconsistent, reducing logging friction genuinely helps — self-monitoring research found monitored groups exercised about 2.07x/week versus 1.36x for controls, with a ~7.8% predicted VO2max bump (PubMed 2489846). If you already train four days a week like clockwork, you mostly just want the data correct. Set your expectations by which lifter you are.

FAQ

Does the Apple Watch count reps automatically?

No. As of 2026, Apple's native strength workout modes log only duration, heart rate, and calories. There is no built-in rep, set, or exercise detection — you need a third-party app like Riven, Motra, Gymatic, Rep Up, or Fitnexx.

How accurate are automatic rep counter apps?

Better than they used to be, but not solved. Independent testing rated the category leader 3/5, and user reviews across apps report frequent manual correction — especially on cables, machines, and odd wrist positions. Riven's internal testing showed exact counts on ~83% of sets and within ±2 reps on 100%.

Why do rep counters fail on cable and machine exercises?

Because a wrist sensor can't see load or joint range of motion directly — it infers reps from wrist movement. When your wrist barely moves (machine holds, many cable lifts) there's almost no signal to count, so every wrist-motion app struggles there.

Do any apps tell me if I reached muscle failure?

Most rep counters only tally reps. Riven also estimates proximity to failure from rep-speed decay (velocity loss) and heart-rate behavior — a signal that maps to how hard the set actually was, not just how many reps you did.

---

Riven counts your reps, names the exercise, and tells you whether you hit failure — all from the Apple Watch you already own. Try it at riven.fit.

Sources

Baraa Bilal
Founder of Riven. Writes about measurement, training, and the small honest signals that separate effort from results.
Read next
Stop Counting Your Reps: Why Manual Rep-Counting Is Quietly Wrecking Your Sets