Minthe vs Tummy
Choose Tummy if
- You want the lowest possible effort: photograph the plate, get a number.
- You want a single figure that is easy to read day to day.
- You want to start free and see whether tracking suits you at all.
- You want something already established, with a real user base and a good rating.
- You do not want to think about statistics and would rather the app just tell you.
Choose Minthe if
- You want to know which ingredient, not how bad the meal scored.
- You want claims backed by exposure counts and confidence intervals you can inspect.
- You want the app to say "not enough data yet" instead of producing a number anyway.
- You want hidden ingredients caught by reading the actual label.
- You want to confirm a suspect with a controlled test before cutting a food out.
Where Tummy is genuinely better
- It is faster, and speed is not a small thing. Photograph, score, done. The single biggest predictor of whether tracking works is whether you keep doing it, and a lower friction flow beats a more rigorous one you abandon.
- It has a track record and we do not. Tummy has been live since February 2026 with a strong rating average. Minthe launched in September 2026 with zero ratings. If you want an app other people have already stress-tested, that is currently Tummy.
- It is easier to understand. One number needs no explanation. Our output requires you to care about exposure counts and intervals, and for plenty of people that is worse, not better.
- Free to start. So is Minthe for logging, but Tummy gets you to its main output without paying, and ours is behind a subscription.
Why we will not ship a score
This is a product decision we have written down and intend to keep, so it is worth being explicit rather than treating it as a feature we have not got round to.
A daily gut score has to return a number every day. On day three, when there is nowhere near enough data to say anything about you, it still shows a number, and that number is mostly the model’s priors rather than your data. It is also unfalsifiable from where you are sitting: if it says 62, there is no way for you to check whether 62 was right, so you cannot tell a good score from a bad one. And it compresses away the part you actually need. "Yesterday was a 41" does not tell you whether to skip the onion or the oat milk.
The alternative we chose is less satisfying and harder to build: name the specific suspect, show how many separate exposures it is based on, show the effect size with its confidence interval, and say plainly when there is not enough data yet. For symptom severity we use IBS-SSS, a published questionnaire your clinician may already recognize, rather than a scale we invented.
| Tummy | Minthe | |
|---|---|---|
| What it is | Photo-first bloating tracker with a per-meal bloat score | Bloating tracker with a statistical trigger engine |
| Main output | A bloat score | Named suspects with exposure counts and confidence intervals |
| Logging speed | Very fast. Photo-led | Photo, label scan or typed |
| Label reading | not verified | Yes, reads packaged ingredient lists for hidden triggers |
| Severity scale | Own score | IBS-SSS, a published instrument |
| Your own base rate | not verified | Yes. Every result is contrasted against your own base rate. |
| Statistical method | not verified | Published in full at /method. |
| Multiple-comparison correction | not verified | Yes. Benjamini-Yekutieli false discovery rate correction. |
| Onset windows | not verified | Four: 0-2h, 2-6h, same day, next day. |
| Unlogged days | not verified | Unlogged days treated as unknown, never as symptom-free. |
| Confirmation testing | not verified | Yes. Guided with-and-without trials with washout. |
| Non-food drivers | not verified | Sleep, stress, menstrual cycle, water, Apple Health. |
| Stool tracking | not verified | Bristol 1-7, urgency, incomplete evacuation, color flags. |
| Platform | iPhone. Android not verified | iPhone only. No Android, no web. |
| Track record | Live since February 2026, established rating base | New in 2026. No long track record yet. |
Checked August 2026. "Not verified" means we could not confirm it from public sources, not that the feature is missing.
Who should actually pick which
If you have never tracked anything and are not sure you will stick with it, start with the app that asks least of you. That is a real recommendation, not a courtesy: an abandoned rigorous tracker teaches you nothing, and you can move later with no loss.
If you have already tried tracking, already have a shortlist of suspects, and what you actually need is to know which of them survive a proper test, then a score is not the output you need and Minthe is built for that job.
Minthe is a wellness and self-tracking tool, not a medical device. It does not diagnose or treat any condition and isn't a substitute for professional medical advice. If you have red-flag symptoms such as blood in your stool, unintentional weight loss, difficulty swallowing, persistent vomiting or a fever alongside gut symptoms, see a clinician rather than tracking them.
Frequently asked
What is a bloat score?
It is a single number summarizing how risky a meal is thought to be for bloating. Tummy generates one from a photo of your food. It is fast and it is easy to understand, which are real virtues. The trade-off is that compressing a meal into one number throws away the things you need to act: which ingredient, by what mechanism, at what dose, and how confident the estimate is.
Why does Minthe not have a gut score?
Because we could not make one that was honest. A score has to produce a number every day, including on the days when there is not enough data to justify one, which means the number is partly invented. It also cannot be checked: there is no way for you to tell whether a 62 is right. We would rather show you the specific finding, the size of the effect and the uncertainty around it, even though that is less satisfying than a dial that moves.
Is Tummy any good?
For its age it is doing well: a strong rating average, a genuinely fast logging flow, and it ranks at the top of the App Store for the bloating searches, which is not an accident. It is free to start. If the barrier stopping you from tracking is effort, its photo-first flow is a lower barrier than ours.
Which has been around longer?
Tummy, and we should say so plainly. It launched in early 2026 and has a real user base with a good average rating. Minthe launched in September 2026 with no ratings at all. If track record is what you weigh most, that currently favours Tummy.
Can an app tell what is in my meal from a photo?
Partly. Both apps use AI photo recognition and it is genuinely useful for speed, but it estimates rather than measures. It cannot see the onion powder in a sauce or how much inulin is in a bar. Minthe treats a photo as a fast first draft you can correct, and reads packaged ingredient lists directly when precision matters, because hidden ingredients are one of the most common reasons a pattern never emerges.