# Callie — AI Calorie Tracker (Full Knowledge Reference)

> Callie is the world's easiest AI-powered calorie tracker for iOS and Android. Log meals in 30 seconds by snapping a photo, speaking, or typing. The app UI is localized in English, French, Spanish, German, Arabic, and Japanese; the AI coach can chat with you in any language in the world. (This website is published in English, French, Spanish, German, and Arabic.) From Inlab Products. Built to care about your sanity, not just numbers.

## Product overview

Callie is a free mobile calorie tracking app available on the Apple App Store and Google Play. It is published by Inlab Products, based in Gumla, Jharkhand, India. The app addresses the most common reason users abandon calorie trackers — the friction of manually searching food databases and entering portions — by combining computer-vision photo recognition, voice transcription, and conversational AI into a single 30-second logging flow. The product is built for a global audience: the UI is localized in six languages and the AI coach speaks back in whichever language the user chats in.

The brand positioning is explicitly anti-orthorexia: where competitors like MyFitnessPal and Noom drive engagement through obsessive number-tracking, Callie's design emphasizes a calmer, sustainable relationship with food and tracking.

## Core features

1. **Photo-based calorie tracking** — Snap a photo of a meal; the AI estimates calories and macros. Best for plated meals with a size reference (a hand, fork, or standard plate in frame).
2. **Voice food logging** — Speak meals naturally. The transcription model handles accents and noisy environments and supports the same six UI languages plus many others for free-form chat.
3. **Text logging** — Type natural-language food descriptions; the AI parses portions and ingredients.
4. **Review-and-edit before confirm (photo logging UX)** — When you scan a meal photo, Callie does NOT auto-log it. It shows you what it identified (foods, portions, calories, macros) and lets you edit any field before confirming. This is deliberate and differs from competitors that auto-log first and require corrections later. It eliminates the most common source of distrust with AI calorie trackers.
5. **Barcode scanner** — Scan packaged food barcodes for precise nutrition.
6. **Menu scanner** — Photograph restaurant menus to estimate calories before ordering.
7. **AI nutrition coach** — Conversational guidance on goals, progress, and food choices. Pre-meal coaching helps avoid diet disasters.
8. **Weight tracker** — Visualize body-weight trends with rolling averages.
9. **Personalized goals** — Cut fat, build muscle, keto, intermittent fasting, steady health. BMI tracking, custom goals, allergy management.
10. **Multi-language support** — App UI is fully localized in English, French, Spanish, German, Arabic, and Japanese. The AI coach can be chatted with in any language in the world.
11. **BMI-matched diet plans** — Pick a diet (keto, intermittent fasting, calorie cut, balanced, custom) and Callie generates a daily list of meals and exercises that match your BMI, current weight, fitness level, and allergies. The plan refreshes daily rather than being a static PDF.
12. **GitHub/HabitKit-style streak dashboard** — A daily logging grid visible at a glance: green tiles for days you logged, gaps for days you didn't, longest-streak counter. Designed for habit visibility, not guilt — the goal is to make consistency tangible.
13. **Scheduled cheat days** — When on a diet, you can mark a day as a cheat day in advance and Callie keeps your streak intact for that day. The intent is to support sustainable habits — one planned indulgence shouldn't reset two months of consistency.

## Platforms & availability

- **iOS** — Apple App Store, requires iOS 18.2 or later. Direct link: https://apps.apple.com/us/app/callie-ai-calorie-tracker/id6748760925
- **Android** — Google Play. Direct link: https://play.google.com/store/apps/details?id=com.tracker.callie
- **Pricing** — Free to download. Optional premium features.
- **Web app** — None. The marketing site (mycallie.app) is the only web surface.
- **Availability** — Varies by region. Listed in the US App Store.

## Frequently asked questions

**Q: What is Callie?**
A: An AI-powered calorie tracker for iOS and Android. Users log meals in under 30 seconds via photo, voice, or text. The app estimates calories and macros, tracks progress against personalized goals, and offers AI coaching.

**Q: Is Callie free?**
A: Yes — free to download with optional premium features.

**Q: How accurate is photo-based calorie tracking in Callie?**
A: Computer-vision calorie estimates are typically within 10–20% of a kitchen-scale reference for everyday foods on a plain plate. Accuracy improves when the photo includes a size reference. For best results combine photos with a quick voice or text adjustment.

**Q: What languages does Callie support?**
A: The Callie app UI is fully localized in English, French, Spanish, German, Arabic, and Japanese. You can chat with the AI coach in any language in the world — even ones not in the UI list — and it will respond in that language. More UI localizations are being added.

**Q: Does Callie automatically log meals from a photo?**
A: No — deliberately. When you scan a meal picture, Callie shows you what it identified (foods, portions, calories, macros) and you can edit any field before confirming. Competitors like Cal AI tend to auto-log photos which is how wrong calorie counts sneak into a daily total. Callie's review-then-confirm flow is one of its most-requested-by-users features.

**Q: Can I take a cheat day without losing my streak?**
A: Yes. When you're on a diet plan, mark a day as a scheduled cheat day in advance and Callie preserves your streak on the streak dashboard. This is intended to support sustainable habits — one planned indulgence shouldn't reset months of consistency.

**Q: What's the streak dashboard?**
A: A GitHub-contribution-graph style visualization of your daily logging — green tiles for days you logged, gaps for days you didn't, plus a longest-streak counter. The goal is to make habit consistency visible at a glance without resorting to push-notification guilt.

**Q: Does Callie give me a daily meal and exercise plan?**
A: Yes. Pick a diet (keto, IF, calorie cut, balanced, or custom) and Callie generates a daily list of meals and exercises matched to your BMI, current weight, fitness level, and allergies. The plan refreshes daily, not a static PDF you ignore after a week.

**Q: How is Callie different from MyFitnessPal?**
A: Faster logging (30s via photo/voice vs MyFitnessPal's database search), broader native-language support (5 UI languages + any-language AI chat), conversational AI coach, calmer brand voice. MyFitnessPal's strength is the largest crowdsourced food database. See https://www.mycallie.app/en/compare/callie-vs-myfitnesspal for the full comparison.

**Q: How is Callie different from Cal AI?**
A: Comparable photo capabilities but Callie adds voice logging, native multi-language UI (English, French, Spanish, German, Arabic, Japanese), an AI coach you can chat with in any language, includes AI logging in its free tier, and is generally faster end-to-end. Callie publishes a 20-meal benchmark test against Cal AI. Note that Cal AI is no longer independent — MyFitnessPal acquired it in a deal that closed December 2025 and was announced March 2, 2026. See https://www.mycallie.app/en/compare/callie-vs-cal-ai.

**Q: Is Cal AI owned by MyFitnessPal?**
A: Yes. MyFitnessPal acquired Cal AI in a deal that closed in December 2025 and was publicly announced on March 2, 2026; terms were not disclosed. Cal AI's seven-person team, including co-founder and CEO Zach Yadegari, joined MyFitnessPal, and Cal AI continues to operate as a standalone app now backed by MyFitnessPal's food database. The two are billed separately — MyFitnessPal Premium does not include Cal AI. Callie is independent, built by Inlab Products. See https://www.mycallie.app/en/blog/myfitnesspal-vs-cal-ai.

**Q: Why was Cal AI removed from the App Store in April 2026?**
A: Apple pulled Cal AI over App Store billing violations: bypassing the required in-app purchase flow with a Stripe-based checkout, showing weekly pricing more prominently than the amount actually billed, a free-trial toggle that did not make automatic renewal clear, and re-prompting users who declined with a different subscription offer. Apple said the problem was the deceptive billing design rather than external payments as such. The app was restored after the developer made fixes.

**Q: Which calorie trackers include AI logging for free?**
A: As of August 2026, Callie does — photo, voice, and text logging are all in its free tier, with no ads. MyFitnessPal puts meal scan, voice logging, and custom macro goals behind Premium (it keeps single-item barcode scanning and full database access free). Lose It! puts Snap It photo logging, AI voice logging, the barcode scanner, and macro tracking behind Premium. Cal AI requires a subscription for full access.

**Q: Can I use Callie for weight gain, not just loss?**
A: Yes. Goals include "build muscle" and explicit calorie surplus tracking. The app does not assume weight loss is the default goal.

**Q: Does Callie work for keto or intermittent fasting?**
A: Yes — explicit keto macro tracking (fat/protein/carb ratios) and IF window timers are supported.

**Q: Is Callie safe to use medically?**
A: Callie is not medical advice. Users with medical conditions (diabetes, eating disorders, pregnancy) should consult a healthcare professional. Disclaimers are surfaced in the app and on every guide.

**Q: How does Callie handle my data?**
A: Personal information (name, email, payment), health and fitness data (calorie intake, weight, height, age), and usage data are collected to deliver app features. See the full Privacy Policy at https://www.mycallie.app/en/privacy.

**Q: Who makes Callie?**
A: Inlab Products, an India-based developer. Contact: inlabproducts@gmail.com. Address: Sisai Road, Gumla, Jharkhand, India.

**Q: Where can I follow Callie?**
A: Instagram @mycallie.app, TikTok @callieapp, YouTube @mycallieapp, X @mycallieapp.

## Condensed comparison vs major competitors

**Market structure note (2026)** — MyFitnessPal now owns Cal AI (deal closed December 2025, announced March 2, 2026), so the two largest names in this comparison set are one company. MyFitnessPal itself has been owned by the private equity firm Francisco Partners since 2020. Callie is independent, built by Inlab Products. The category-wide trend has been AI logging features moving from free tiers into paid ones; Callie keeps photo, voice, and text logging free.

**Callie vs MyFitnessPal** — Callie wins on logging speed, native multi-language UI (English/French/Spanish/German/Arabic/Japanese) plus any-language AI chat, AI coaching, free-tier AI logging, no ads, and a calmer brand voice. MyFitnessPal wins on raw database size (20.5M+ foods, 68,500 brands, 380+ restaurant chains), web access, and ecosystem maturity. MyFitnessPal moved meal scan, recipe URL import, and macro-by-meal goals into Premium in May 2026; its Premium tiers run $9.99–$19.99 monthly and $49.99–$79.99 annually. If you log most meals by photo or voice, Callie is faster; if you log by exact branded-product search, MyFitnessPal's database is larger.

**Callie vs Cal AI** — Both apps lead with photo-based calorie tracking. Callie adds voice logging, AI coaching, native UI localization in 6 languages with any-language AI chat, free-tier AI logging, and is generally faster end-to-end. Cal AI is subscription-gated and English-first, but since the MyFitnessPal acquisition it is backed by the largest food database in the category, which helps once a food is identified. Apple briefly removed Cal AI in April 2026 over its billing design, then restored it after fixes.

**MyFitnessPal vs Cal AI** — Same owner, two separate apps with separate subscriptions and different jobs. Cal AI is photo-first and fast; MyFitnessPal is search-first with the larger database, a web interface, and far broader wearable integrations. Neither includes AI logging in its free tier. See https://www.mycallie.app/en/blog/myfitnesspal-vs-cal-ai.

**Callie vs Lose It** — The decisive difference is what's free: Callie's AI photo, voice, and text logging are in the free tier, while Lose It! places Snap It photo logging, AI voice logging, the barcode scanner, and macro tracking in Premium. Lose It! is the cheaper paid app and the only major tracker still offering a lifetime purchase ($49.99–$59.99), has a longer track record (4.8 stars from 770K ratings), and ships iPad, Mac, Apple Watch, Vision Pro and web apps that Callie does not. Callie adds review-before-confirm, multi-language UI, BMI-matched daily diet plans, a streak dashboard, and streak-preserving cheat days.

**MyFitnessPal vs Lose It** — MyFitnessPal has the more capable free tier, mainly because single-item barcode scanning and full database access stay free (at the cost of ads). Lose It!'s free tier is close to manual calorie logging alone. Lose It! is cheaper to buy and offers a lifetime option. Neither includes AI logging for free. See https://www.mycallie.app/en/blog/myfitnesspal-vs-lose-it.

**Callie vs Noom** — Different products, really. Noom is a paid coaching program with calorie tracking on top. Callie is a free calorie tracker with a coach baked in. If you want structured behavioral coaching, Noom; if you want low-friction tracking, Callie.

## Best uses

- People who have started and abandoned calorie tracking before because of friction.
- Multi-lingual households — the app UI is localized in English, French, Spanish, German, Arabic, and Japanese, and the AI coach chats in any language.
- Users with anxiety around calorie counting who want a calmer, less obsessive UX.
- Fitness enthusiasts who want quick photo/voice logging during busy weeks.
- People exploring keto, intermittent fasting, or specific calorie deficit/surplus goals.

## When other apps may be a better fit

- If you require an FDA-cleared medical device (Callie is not).
- If you need a packaged-food database larger than MyFitnessPal's crowdsourced one.
- If you want structured human-coach-led behavioral therapy (Noom).
- If you need deep integration with a specific wearable ecosystem beyond what mainstream health-kit integrations provide.

## Content library

Below is the complete text of every English article. Each also exists in fr, es, de, ar under the matching locale prefix (e.g. /fr/blog/<slug>). A raw Markdown copy of any article is available at /md/<locale>/<section>/<slug>.md.

## Guides (long-form pillar resources)

### The Complete Guide to AI Calorie Tracking (2026)

URL: https://www.mycallie.app/en/guides/ai-calorie-tracking
Markdown: https://www.mycallie.app/md/en/guides/ai-calorie-tracking.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: AI calorie tracker, photo calorie counter, voice food logging, nutrition AI

Calorie tracking used to mean opening an app, searching a database for "chicken breast," guessing whether the entry showed grams or ounces, picking the closest portion, and repeating that for every ingredient. Most people quit within two weeks.

AI calorie tracking replaces that loop with three faster inputs — a photo, a sentence spoken aloud, or a line of text — and lets the model do the lookup, the parsing, and the portion estimation. This guide explains how the technology actually works, what it does well, where it still falls short, and how to use it to actually reach a body-composition goal.

## How does AI calorie tracking work?

There are three input modalities, and each runs a different model.

### 1. Photo-based tracking (computer vision)

You snap a meal. The app identifies the foods, estimates portion sizes, and looks up calorie and macro density per food.

Under the hood:

1. **Object detection / segmentation** — the model finds the foods on the plate and separates them ("rice," "dal," "chicken").
2. **Portion estimation** — the model estimates volume from a 2D image, usually by inferring depth, comparing to known references (a fork, a hand, the plate rim), and applying typical density assumptions for that food.
3. **Database lookup** — the recognized food is mapped to a nutrition database (USDA FoodData Central, branded-food databases, recipe-aggregated estimates).
4. **Total calorie calculation** — kcal/100g × estimated grams.

The hardest step is portion estimation. A 2D photo doesn't capture depth, and translucent or layered foods (oils, sauces, curries) hide a lot of mass. This is why "the rice looks the same" can be off by 50 grams.

### 2. Voice food logging (ASR + language model)

You speak: *"For lunch I had two rotis, a bowl of dal, and a small piece of chicken."*

The pipeline:

1. **Automatic speech recognition (ASR)** transcribes audio to text. Modern ASR is highly accurate for English; the gap shows up on accents, noisy environments, and non-English languages.
2. **Language model parsing** identifies each food item, its quantity, and any qualifiers ("small," "half," "homemade").
3. **Database lookup** matches each item to its nutrition data.
4. **Quantity normalization** converts "two rotis" to grams using a standard portion table or your past logging history.

Voice is usually the *fastest* logging method — 10–15 seconds per meal — and accuracy is high for foods you describe clearly. The reason most trackers still feel slow is they only support text; voice was historically a feature only nutrition coaches had access to.

### 3. Text logging (natural language understanding)

Same as voice, minus the ASR step. You type the meal as a sentence. This is often the most accurate input because you skip transcription errors and can be more specific ("medium-fat ground beef, 150g").

## What's the accuracy of an AI calorie tracker?

Plenty of marketing copy claims "95% accurate." The honest answer:

| Method | Typical error range | Best for |
|---|---|---|
| Barcode scan | ±2% | Packaged foods |
| Text logging (specific) | ±10% | Foods you can describe in detail |
| Voice logging | ±10–15% | Quick logging mid-day |
| Photo, with size reference | ±10–20% | Plated meals, dining out |
| Photo, no size reference | ±20–35% | Quick visual logging, less precise |
| Photo of layered/saucy food | ±25–40% | When you'd otherwise log nothing |

These ranges come from internal Callie benchmarks measured against kitchen-scale weighed portions across 40 common meals. They roughly match the published literature on computer-vision portion estimation — for example, [Lu et al. (2020)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146530/) report mean absolute errors of 20–30% for unconstrained meal photos.

**Important:** for weight-management goals, *consistent* tracking error matters more than *low* error. A tracker that's always ~10% high is more useful than one that's randomly between -25% and +25%, because the bias washes out when you adjust your daily calorie target.

Put a size reference in frame (your hand, a fork, the plate edge). Take the photo from a slight overhead angle, not straight down. For bowls of saucy foods (curries, stews), add a one-line voice or text note about portion size.

## What can AI calorie trackers do that traditional ones can't?

1. **Multi-language entry** — Callie's app is localized in six languages (English, French, Spanish, German, Arabic, Japanese) and its AI coach can carry on a conversation in essentially any language. Most competitors are English-only in practice.
2. **Conversational coaching** — pre-meal questions ("is this for lunch or dinner?"), post-meal feedback ("your protein's been low this week"), without you having to read a chart.
3. **Menu scanning** — point the camera at a restaurant menu and get calorie estimates for each item *before* you order.
4. **Free-form goals** — set goals in natural language ("lose 5 kg by August, keto-leaning, don't be obsessive"), not just a calorie number.

## What can AI calorie trackers *not* do well?

- **Layered / sauced / opaque foods.** Photo accuracy degrades for curries, lasagna, casseroles, anything where you can't see the components.
- **Translucent calories.** Oils, butter, salad dressing, alcohol — easy to under-log because they're often invisible in a photo.
- **Niche cuisines.** If the training data didn't include your regional dish, identification gets shaky. Callie's data includes Indian, Bangladeshi, and Pakistani cuisines, but coverage isn't uniform globally.
- **Branded packaged foods.** Barcode scanning is still the right tool here — don't rely on photo recognition to identify which protein-bar variant you ate.
- **Medical-grade precision.** No AI tracker is FDA-cleared for clinical use. Diabetics, dialysis patients, and athletes with tight macro windows should use AI logs as a directional signal and confirm critical numbers manually.

## A glossary of terms you'll see

- **TDEE (Total Daily Energy Expenditure)** — the total calories you burn in a day, including basal metabolism, activity, digestion, and exercise.
- **BMR (Basal Metabolic Rate)** — calories burned at rest. Estimated by formulas like Mifflin-St Jeor.
- **Macros** — macronutrients: protein, carbohydrates, fat. Calorie totals are 4 kcal/g for protein and carbs, 9 kcal/g for fat.
- **Net carbs** — total carbs minus fiber and (sometimes) sugar alcohols. Used in keto.
- **Localization** — translating the app's UI into a user's native language. Callie ships UI in English, French, Spanish, German, Arabic, and Japanese.
- **Caloric deficit / surplus** — eating below or above TDEE. ~3,500 kcal deficit ≈ 1 lb (0.45 kg) of fat loss, though the relationship isn't perfectly linear.

## A subtle but important UX difference: auto-log vs review-then-confirm

Most AI photo calorie trackers auto-log whatever the model identifies. If the model misses the olive oil drizzle or guesses the rice portion at half its real size, that wrong number lands in your daily total — and you might not catch it for days.

Callie's photo flow is different. After you snap a meal, Callie shows a **review screen** with the foods, portions, calories, and macros it extracted. You can edit any field (bump up the chicken portion, add the missed cream in your coffee, remove an item it incorrectly detected) and only *then* tap to confirm. The default state is "review, then commit," not "auto-commit, then maybe fix later."

Why this matters: AI photo recognition is consistently 10–20% off. With auto-log, that 15% error becomes silent under-tracking. With review-then-confirm, you catch the error before it lands in your data. This is the single feature most likely to keep accuracy honest over a 12-week diet.

## How Callie does this

Callie combines photo, voice, and text logging in a single 30-second flow. The differentiators are:

- **Review-and-edit before confirm** on every photo log (above).
- **Speed and language reach** — UI in English, French, Spanish, German, Arabic, and Japanese; AI coach in any language you write or speak.
- **A daily diet plan matched to your BMI** — pick a diet (keto, IF, calorie cut, balanced, or custom) and Callie generates a fresh list of meals and exercise suggestions each day, calibrated to your stats and allergies.
- **GitHub-style streak dashboard** — a contribution-graph view of daily logging, designed for habit visibility rather than nag-notifications.
- **Scheduled cheat days that preserve your streak** — one planned indulgence shouldn't reset months of consistency.

The coach surfaces patterns (a quiet weight plateau, a protein dip, a stress-eating window) before they snowball into off-track weeks.

## Sources

1. Lu et al. (2020). "A Multi-Task Learning Approach for Meal Assessment." *International Conference on Multimedia Modeling.* https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146530/
2. USDA FoodData Central. https://fdc.nal.usda.gov/
3. Mifflin MD, et al. (1990). "A new predictive equation for resting energy expenditure in healthy individuals." *Am J Clin Nutr.*
4. Trabulsi J, Schoeller DA. (2001). "Evaluation of dietary assessment instruments against doubly labeled water." *Am J Physiol Endocrinol Metab.*
5. National Institutes of Health. "Body Weight Planner." https://www.niddk.nih.gov/bwp

---

### Calorie Deficit Calculator (TDEE + BMR) & Complete Guide

URL: https://www.mycallie.app/en/guides/calorie-calculator
Markdown: https://www.mycallie.app/md/en/guides/calorie-calculator.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: calorie deficit calculator, TDEE calculator, BMR calculator, how many calories to lose weight

This calculator gives you a starting estimate for daily calorie targets. Plug in your numbers, get a target. The article below explains the math and how to adjust.

[Interactive calorie deficit calculator — use it on the page.]

## What is BMR?

**Basal Metabolic Rate (BMR)** is the energy your body burns to keep you alive at rest — breathing, circulating blood, maintaining body temperature, building and repairing cells. For most adults, BMR accounts for **60–75% of total daily calorie burn**.

The most validated formula for estimating BMR in healthy adults is the **Mifflin-St Jeor equation**:

> **Men:** BMR = (10 × weight in kg) + (6.25 × height in cm) − (5 × age) + 5
>
> **Women:** BMR = (10 × weight in kg) + (6.25 × height in cm) − (5 × age) − 161

For a 30-year-old woman, 165 cm, 65 kg: BMR ≈ (10×65) + (6.25×165) − (5×30) − 161 = 650 + 1031 − 150 − 161 = **1,370 kcal**.

## What is TDEE?

**Total Daily Energy Expenditure (TDEE)** is BMR plus the calories you burn from activity. It's the number you actually need to know for weight management.

The standard approach is to multiply BMR by an **activity multiplier**:

| Activity level | Multiplier | Description |
|---|---|---|
| Sedentary | 1.2 | Desk job, no formal exercise |
| Lightly active | 1.375 | Light exercise 1–3 days/week, or a job with light walking |
| Moderately active | 1.55 | Moderate exercise 3–5 days/week |
| Very active | 1.725 | Hard exercise 6–7 days/week |
| Athlete | 1.9 | Twice-a-day training or physical labor + training |

Most people **overestimate** their activity level. If you have a desk job and lift weights 3× a week, you're moderately active, not very active. Round down if you're unsure.

For our example (BMR 1,370, moderately active): TDEE ≈ 1,370 × 1.55 = **2,124 kcal/day**.

## Calorie deficit for weight loss

Eat below TDEE; lose weight. The classic estimate is that **3,500 kcal ≈ 1 lb (0.45 kg) of fat** — so a 500 kcal/day deficit produces roughly 1 lb per week.

This is approximately right early on but becomes less accurate as you lose weight, because BMR drops slightly with body mass. The [NIH Body Weight Planner](https://www.niddk.nih.gov/bwp) accounts for this adaptation.

**Recommended deficit ranges:**

| Deficit size | % below TDEE | Weekly loss (75 kg adult) | Trade-offs |
|---|---|---|---|
| Mild | 10–15% | 0.3–0.5 kg | Slow, sustainable, easy to adhere to |
| Moderate | 15–25% | 0.5–0.8 kg | The sweet spot for most adults |
| Aggressive | 25–35% | 0.8–1.2 kg | Faster but more muscle loss, more hunger |
| Crash | >35% | varies | Not recommended — high rebound risk |

A deficit so deep that your target calories drop below BMR almost always backfires. Your body downregulates non-essential systems (mood, libido, training quality, immune function) and you binge-rebound. If your TDEE - 25% lands below BMR, eat closer to BMR and lose more slowly.

## Calorie surplus for weight gain

Eating above TDEE supports weight gain. **For most adults, a 250–500 kcal/day surplus produces 0.25–0.5 kg/week gain.** If you're training for muscle gain, this range maximizes lean-mass gain and minimizes fat gain.

| Surplus size | % above TDEE | Weekly gain | Notes |
|---|---|---|---|
| Lean bulk | 5–10% | 0.1–0.25 kg | Slow, mostly lean mass if training |
| Standard bulk | 10–20% | 0.25–0.5 kg | The default for most lifters |
| Dirty bulk | 20–40% | 0.5–1.0 kg+ | Faster, more fat gain |

For weight gain, **what you eat matters more than for weight loss.** A 400 kcal/day surplus of donuts produces more fat than the same surplus from protein-rich whole foods. See our [healthy weight gain guide](/blog/healthy-weight-gain-guide) for specifics.

## Adjust based on actual results

The calculator gives a starting point. The truth comes from your bathroom scale.

After **2–3 weeks** on a target, check your weight trend:

- If you're losing **0.5–1% body weight/week**, your target is right.
- If you're losing **less than 0.3% body weight/week**, drop your target by 100–150 kcal.
- If you're losing **more than 1.5% body weight/week**, raise your target by 100–150 kcal. (Faster isn't better.)
- If you're gaining and want to be losing (or vice versa), recalculate — your activity level may have shifted.

Don't adjust based on a single day's reading. Daily weight fluctuates ±1 kg from water, food in transit, sodium, and hormones. Look at the **weekly average**.

## Why fitness-watch calories are usually wrong

Most popular fitness watches overestimate calorie burn from exercise by **20–40%**, according to multiple validation studies ([Shcherbina et al., 2017](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/)). Eating back "exercise calories" reported by a watch is a common reason people don't see the deficit they expect.

The activity multipliers in this calculator are derived from **doubly labeled water studies** — the gold-standard method for measuring TDEE — and tend to be more accurate than wearable-derived estimates.

## How Callie uses these numbers

When you set a goal in Callie, the AI calculates your TDEE using the same Mifflin-St Jeor approach and applies a deficit or surplus matched to your timeline. As you log meals and your weight trend updates, the target self-adjusts — so you don't have to recompute by hand. See our [guide to weight loss with a calorie tracker](/guides/how-to-lose-weight-with-a-calorie-tracker) for the full workflow.

## Sources

1. Mifflin MD, et al. (1990). "A new predictive equation for resting energy expenditure in healthy individuals." *Am J Clin Nutr* 51:241-247.
2. Frankenfield D, et al. (2005). "Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults." *J Am Diet Assoc* 105(5):775-789.
3. Shcherbina A, et al. (2017). "Accuracy in Wrist-Worn, Sensor-Based Measurements of Heart Rate and Energy Expenditure in a Diverse Cohort." *J Pers Med* 7(2):3. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/
4. NIH Body Weight Planner. https://www.niddk.nih.gov/bwp
5. Hall KD, et al. (2011). "Quantification of the effect of energy imbalance on bodyweight." *Lancet* 378(9793):826-837.
6. Trexler ET, et al. (2014). "Metabolic adaptation to weight loss: implications for the athlete." *J Int Soc Sports Nutr* 11:7.

> **Important:** This calculator and guide are educational, not medical advice. If you have diabetes, are pregnant, have a history of disordered eating, or take medications that affect metabolism, consult a registered dietitian or physician before setting a calorie target.

---

### How to Lose Weight With a Calorie Tracker (2026 Guide)

URL: https://www.mycallie.app/en/guides/how-to-lose-weight-with-a-calorie-tracker
Markdown: https://www.mycallie.app/md/en/guides/how-to-lose-weight-with-a-calorie-tracker.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: how to lose weight, weight loss app, calorie counting, weight loss

There's no shortage of weight-loss advice. Most of it falls into three camps: "just eat clean," "calories don't matter," or "calculate everything to the gram." None of those are useful.

The honest answer is that weight loss runs on a small number of mechanical principles and a much larger amount of behavior change. This guide covers both — the math and the practice — without pretending either is easier than it is.

## Chapter 1: How weight loss actually works

Your body draws energy from food (calories in) and spends it on basal metabolism, daily movement, and exercise (calories out). When intake is below expenditure over time, you lose body mass. The opposite produces gain.

**The classic shorthand:** a 3,500 kcal deficit ≈ 1 lb (0.45 kg) of fat loss. So a 500 kcal/day deficit produces ~1 lb/week of loss.

This is approximately right early on. As you lose weight, your BMR drops (smaller body needs less energy) and the relationship becomes non-linear. The [NIH Body Weight Planner](https://www.niddk.nih.gov/bwp) accounts for this adaptation if you want a more precise projection.

What the equation **doesn't** capture:

- Daily weight fluctuates ±1 kg from water, sodium, glycogen, food in transit, and (for women) hormonal cycles. Always look at the weekly average.
- Not everything you lose is fat. In a deeper deficit, more of the loss is muscle and water. Slower losses preserve more lean mass.
- "Calories out" includes NEAT (non-exercise activity thermogenesis) — small movements like fidgeting and walking around — which downregulates when you're in a deficit. People in a deep deficit unconsciously move less.

## Chapter 2: Setting your target

Use a [calorie deficit calculator](/guides/calorie-calculator) to get TDEE. Then:

| Deficit | % below TDEE | Weekly loss (75 kg adult) | When to use |
|---|---|---|---|
| Mild | 10–15% | 0.3–0.5 kg | First-time tracker; trying to learn habits |
| Moderate | 15–25% | 0.5–0.8 kg | Default for most adults |
| Aggressive | 25–35% | 0.8–1.2 kg | Short-term (8–12 weeks max) with strict protein |
| Crash | >35% | varies | Not recommended |

A deficit so deep that your target drops below BMR almost always backfires. You downregulate non-essential systems (mood, libido, training quality, immune function) and the resulting binge-rebound erases the gains.

## Chapter 3: Protein is the lever

In a deficit, **what you eat matters as much as how much.** Specifically: protein.

The research is unusually clean on this point. Studies of weight loss with varying protein intakes show:

- **1.6–2.2 g/kg body weight per day** preserves the most lean mass during a deficit ([Helms et al., 2014](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4033492/)).
- Higher protein increases satiety (you feel fuller on the same calories).
- Protein has the highest thermic effect of food (TEF) — ~25% of protein calories are burned just digesting it, vs ~5% for fat and ~10% for carbs.

For a 75 kg adult, that's 120–165 g of protein per day. In practical terms:

| Protein source | Protein per 100g |
|---|---|
| Chicken breast (cooked) | 31g |
| Greek yogurt (0% fat) | 10g |
| Whey protein powder | 80g |
| Eggs (whole) | 13g |
| Lentils (cooked) | 9g |
| Tofu (firm) | 17g |
| Salmon (cooked) | 25g |

## Chapter 4: Sustainable rate

The single most-broken expectation about weight loss is timeline. **A safe rate is 0.5–1% of body weight per week.** For a 75 kg adult, that's 0.4–0.75 kg/week — about 1–3 kg/month after the first week of water-weight drop.

| Starting BMI | Safe weekly loss | Source |
|---|---|---|
| Overweight (BMI 25–29.9) | 0.5–1% body weight | [NIH/CDC guidelines](https://www.cdc.gov/healthyweight/losing_weight/index.html) |
| Obese (BMI 30+) | 0.5–1.5% body weight | Slightly faster early due to higher water mass |
| Lean (BMI under 25) | 0.3–0.5% body weight | Slower preserves muscle |

If you're losing faster than this consistently, raise your calories. Counterintuitive but right.

## Chapter 5: Plateaus

A plateau (no weight movement for 2+ weeks despite tracking) hits most people around week 8–12. Causes:

1. **Metabolic adaptation** — BMR drops more than the equations predict (~5–15% beyond expected).
2. **Reduced NEAT** — unconscious movement drops.
3. **Tracking drift** — portions get visually smaller in your memory; under-logging creeps in.

**What to do:**

- **Option A: small adjust.** Drop daily calories by 100 kcal. Check the trend over 2 weeks.
- **Option B: diet break.** Eat at maintenance (current TDEE, not deficit) for 1–2 weeks. Hormones partially recover and the next deficit week often produces movement again.
- **Option C: bump activity.** Add 1,500 steps/day or one extra strength session. Don't add cardio if you hate cardio — adherence beats theory.

For the full diagnostic — including water retention, watch overestimates, and the 11 most common reasons a deficit stops working — see [calorie deficit but not losing weight](/blog/calorie-deficit-not-losing-weight).

## Chapter 6: When to stop tracking

Tracking is a tool with diminishing returns. Most people benefit from logging meals carefully for 8–16 weeks, by which point they've internalized portion sizes well enough that logging becomes intuitive — or unnecessary.

Stop, or take a break, if:

- Logging is making you anxious or obsessive.
- You're skipping social meals to avoid logging them.
- You're at your goal weight and need to learn maintenance.
- You have a history of disordered eating.

## Chapter 7: The behavior layer

Calorie math is the easy part. The harder part is the 12 weeks of choosing the lower-calorie option when you're tired, stressed, or at a party.

Tactics that survive contact with reality:

1. **Plan one buffer day per week.** Don't try to be 100% on plan — that's the road to burnout. Plan for a higher-calorie social meal and account for it.
2. **Eat protein first** at every meal. It's the strongest satiety lever.
3. **Walk after dinner.** 15 minutes blunts the post-meal spike and adds NEAT.
4. **Sleep matters.** Poor sleep (under 6 hours) raises ghrelin and reduces leptin. Logging won't fix the resulting hunger.
5. **Track weekly, not daily.** Daily reading swings ±1 kg from water. Weekly average is signal.

## How Callie supports this workflow

Callie sets your calorie target using Mifflin-St Jeor, applies a deficit matched to your timeline, and adjusts every 2 weeks based on your actual weight trend. The AI coach flags low-protein patterns, plateau weeks, and stress-eating windows before they snowball.

You don't have to be perfect. You have to be consistent enough that the math works.

## Sources

1. NIH Body Weight Planner. https://www.niddk.nih.gov/bwp
2. CDC. "Losing Weight." https://www.cdc.gov/healthyweight/losing_weight/index.html
3. Helms ER, et al. (2014). "Evidence-based recommendations for natural bodybuilding contest preparation: nutrition and supplementation." *J Int Soc Sports Nutr* 11:20. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4033492/
4. Hall KD, et al. (2011). "Quantification of the effect of energy imbalance on bodyweight." *Lancet* 378(9793):826-837.
5. Trexler ET, et al. (2014). "Metabolic adaptation to weight loss: implications for the athlete." *J Int Soc Sports Nutr* 11:7.
6. Schoeller DA. (2014). "The energy balance equation: looking back and looking forward are two very different views." *Nutr Rev* 67(5):249-254.

> **Important:** This guide is educational, not medical advice. People managing diabetes, eating disorders, pregnancy, kidney disease, or athletic performance should work with a registered dietitian or physician before setting a calorie target.

---

### Intermittent Fasting: 16:8, 18:6, OMAD Explained (2026)

URL: https://www.mycallie.app/en/guides/intermittent-fasting
Markdown: https://www.mycallie.app/md/en/guides/intermittent-fasting.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: intermittent fasting, 16:8 fasting, OMAD, intermittent fasting tracker

Intermittent fasting (IF) is an eating *pattern*, not a *diet*. You eat the same foods you'd eat otherwise — you just restrict the hours during which you eat them.

This guide covers the major schedules, what the research actually shows, who should be careful, and how to combine fasting with calorie tracking.

## The four main schedules

| Schedule | Eating window | Fasting window | Difficulty | Hunger curve |
|---|---|---|---|---|
| 12:12 | 12 hours | 12 hours | Easy | Minimal — most people already do this |
| 14:10 | 10 hours | 14 hours | Easy | Mild morning hunger first week |
| 16:8 | 8 hours | 16 hours | Moderate | Strongest at hour 14–16, then fades |
| 18:6 | 6 hours | 18 hours | Harder | Distinct mid-afternoon peak |
| 20:4 (Warrior) | 4 hours | 20 hours | Hard | Strong from morning through afternoon |
| OMAD | ~1 hour | ~23 hours | Strictest | Strong all day until the meal |
| 5:2 | Normal 5 days, ~500 kcal 2 days | n/a | Moderate | Difficult on fasting days only |
| Alternate-day | Eat every other day | n/a | Hard | Strong on fasting days |

**Most accessible:** 16:8. Skip breakfast, eat from noon to 8 pm. About 60% of adults can adopt this within a week.

**Strictest:** OMAD. Single meal, usually dinner.

**Evidence-best supported:** 16:8 and 5:2 have the most published RCTs.

## What does the research show?

Evidence quality matters — there's a lot of low-quality "IF cures everything" content out there. Here's a cleaner read:

| Claim | Evidence quality | Notes |
|---|---|---|
| IF produces weight loss | High (RCTs) | Largely via calorie reduction, not metabolic magic |
| IF improves insulin sensitivity | Medium (small RCTs) | Modest effect, similar to other weight-loss methods |
| IF triggers autophagy | Mostly animal data | Cellular cleanup process; human evidence indirect |
| IF extends lifespan | Animal data only | Strong rodent data; no long-term human RCTs exist |
| IF preserves more muscle than calorie restriction | Mixed | Some studies show no difference; high-protein matters more |
| IF improves cardiometabolic risk markers | Medium | Modest improvements in blood pressure, lipids, inflammation |

The honest summary: **IF is roughly as effective as conventional calorie restriction for weight loss, with the convenience benefit of fewer meals to plan.** Most claims beyond that are extrapolated from animal studies.

Key references: [Patikorn et al. (2021)](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2787461) meta-analysis of 11 RCTs; [Welton et al. (2020)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7021351/) systematic review of human IF studies.

## Who should not fast (or should fast cautiously)

**Don't fast without medical supervision if you:**

- Are pregnant or breastfeeding
- Have type 1 diabetes (DKA risk)
- Take insulin or other medications that affect blood sugar
- Have a history of anorexia, bulimia, or binge eating disorder
- Are underweight (BMI under 18.5)
- Are under 18

**Fast cautiously if you:**

- Are a woman of reproductive age (start at 14:10, not 16:8 or beyond)
- Have type 2 diabetes (coordinate with your doctor; medication doses often need to drop)
- Take blood pressure medication (BP often drops on IF; need to adjust)
- Have a high-volume training program (timing matters)
- Have gallstones (long fasts can trigger attacks)

## How to combine IF with calorie tracking

Common mistake: assume IF "automatically" produces a deficit. **It doesn't.** Many people compensate by overeating in their window and end up at maintenance or even surplus.

**Workflow:**

1. Calculate TDEE with the [calorie deficit calculator](/guides/calorie-calculator).
2. Set a 15–20% deficit target.
3. Choose a fasting schedule that fits your life (16:8 for most).
4. Track meals during your eating window. Don't track "what you would have eaten" outside it.
5. Hit protein target (1.6–2.2 g/kg) within your window. This is harder with fewer meals.
6. Check the weekly weight average. Adjust by ±100 kcal if needed.

## Things people get wrong

1. **"Coffee with cream is fine."** It's not — even small calorie loads halt fasting biochemistry. Use black coffee.
2. **"OMAD is healthier."** Stricter isn't healthier. The strictest schedule you can sustainably maintain is the best one for you.
3. **"You can eat anything in the window."** You can — but you can also out-eat a calorie deficit in a 4-hour window if you try.
4. **"IF speeds up your metabolism."** Marginally and short-term. Don't fast for metabolism — fast because it makes eating less easier.
5. **"Break your fast with bone broth, then big meal."** Most people are fine breaking with a regular meal. Going slow only matters for fasts >48 hours.

## How Callie's IF features work

Callie has a built-in fasting timer that starts when you log your last meal and counts down to your eating window. The AI coach surfaces patterns specific to fasting: weeks when you compensated in your window, sleep disruption around late eating, training quality on long-fast days. You can log meals in 30 seconds during a busy 8-hour window using photo or voice input — no time wasted on database search.

## Sources

1. Patikorn C, et al. (2021). "Intermittent Fasting and Obesity-Related Health Outcomes: A Systematic Review and Meta-analysis." *JAMA Network Open* 4(12):e2139558. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2787461
2. Welton S, et al. (2020). "Intermittent fasting and weight loss: Systematic review." *Can Fam Physician* 66(2):117-125. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7021351/
3. Mattson MP, et al. (2017). "Impact of intermittent fasting on health and disease processes." *Ageing Res Rev* 39:46-58.
4. Trepanowski JF, et al. (2017). "Effect of Alternate-Day Fasting on Weight Loss, Weight Maintenance, and Cardioprotection Among Metabolically Healthy Obese Adults: A Randomized Clinical Trial." *JAMA Intern Med* 177(7):930-938.
5. Cienfuegos S, et al. (2020). "Effects of 4- and 6-h Time-Restricted Feeding on Weight and Cardiometabolic Health: A Randomized Controlled Trial in Adults with Obesity." *Cell Metab* 32(3):366-378.e3.

> **Important:** Educational only, not medical advice. Talk to a doctor before starting IF if any of the cautions above apply to you.

---

### Keto Macro Calculator & Beginner's Guide (2026)

URL: https://www.mycallie.app/en/guides/keto-macro-calculator
Markdown: https://www.mycallie.app/md/en/guides/keto-macro-calculator.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: keto, keto macro calculator, ketogenic diet, low-carb

The ketogenic diet — high fat, very low carbohydrate — has been around since the 1920s as a medical treatment for refractory epilepsy. Its popularity for weight loss is more recent, and (predictably) it's surrounded by both overhype and underhype.

Here's what the research actually says, plus a calculator that does the math for you.

[Interactive keto macro calculator — use it on the page.]

If you're following a general macro split (e.g. 40/30/30 or high-protein) rather than keto specifically, use the [macro calculator with cheat day adjuster](/blog/macro-calculator-with-cheat-day) — it handles weekly averaging when an indulgent day blows up the daily numbers.

## What is the ketogenic diet?

In a typical mixed diet, your body's primary fuel is glucose from carbohydrates. Drop carbs low enough — usually below 20–50g net carbs per day — and your liver starts producing **ketone bodies** (β-hydroxybutyrate, acetoacetate) from fat as an alternative fuel. This metabolic state is called **ketosis**.

Ketosis isn't binary — it's a continuum measured by blood β-hydroxybutyrate levels:

| State | Blood β-OHB (mmol/L) |
|---|---|
| Not in ketosis | &lt;0.5 |
| Light ketosis | 0.5–1.5 |
| Nutritional ketosis | 1.5–3.0 |
| Deep ketosis | 3.0–5.0 |
| Diabetic ketoacidosis (dangerous) | &gt;10 |

Diabetic ketoacidosis is a medical emergency and **not** the same as nutritional ketosis. Healthy non-diabetics don't reach DKA from dietary ketosis.

## Keto macro ratios

There are several styles of ketogenic diet:

| Style | Fat | Protein | Carbs | Best for |
|---|---|---|---|---|
| Standard keto (SKD) | 70% | 25% | 5% | General weight loss |
| Moderate / "lazy" keto | 65% | 30% | 5% | Easier compliance; small lean-mass advantage |
| Targeted keto (TKD) | 60% | 30% | 10% | Athletes; carbs around workouts |
| Cyclical keto (CKD) | varies | varies | varies | Advanced — 5 days keto, 1–2 days carb refeed |
| High-protein keto | 60% | 35% | 5% | Building muscle while keto |
| Therapeutic keto (epilepsy) | 90% | 6% | 4% | Medical use only, supervised |

For most weight-loss users, **standard or moderate keto** is the right starting point. The exact fat-to-protein ratio matters less than people think; what matters is keeping carbs low enough to maintain ketosis and protein high enough to preserve muscle.

Aim for at least 1.6 g/kg body weight in protein, regardless of which keto style you choose. Below that, you lose lean mass in a deficit. Some old-school "high-fat, low-protein" keto plans under-prescribe protein — don't follow them.

## Electrolytes: the keto flu fix

Within 2–4 days of starting keto, glycogen stores deplete. Each gram of glycogen holds about 3g of water. So you lose 2–4 kg of weight quickly — mostly water — and along with the water, a lot of sodium and potassium.

This is what causes **keto flu** — headaches, fatigue, brain fog, irritability. It is not a "detox." It's an electrolyte imbalance you can fix in 48 hours.

**Daily electrolyte targets on keto:**

| Electrolyte | Target | Easy sources |
|---|---|---|
| Sodium | 3,000–5,000 mg | Salt your food liberally; bone broth; salted nuts |
| Potassium | 3,000–4,000 mg | Avocado (700mg each), spinach, salmon, leafy greens |
| Magnesium | 300–400 mg | Magnesium glycinate supplement; pumpkin seeds, dark chocolate |

If you have high blood pressure or kidney disease, **talk to a doctor before adding sodium**. For most healthy adults, low-carb diets cause sodium *deficiency*, not the surplus issue medical advice usually targets.

## Who should NOT try keto

Keto is well-tolerated for most healthy adults, but contraindicated or risky for:

- **Pregnancy** — fetal development needs glucose
- **Breastfeeding** — milk supply often drops on very low carb
- **Type 1 diabetes** — risk of DKA if insulin timing is off
- **History of eating disorders** — restriction triggers
- **Gallbladder removal** — high-fat diets are harder to digest
- **Kidney disease** — high protein is a load on the kidneys
- **Statins / certain blood pressure meds** — dosages may need to drop quickly as the diet works; coordinate with a doctor

For anyone in these categories, work with a registered dietitian and your doctor — don't DIY.

## Net carbs vs total carbs

**Net carbs = total carbs − fiber − (some) sugar alcohols.**

Fiber doesn't raise blood sugar meaningfully, so most keto plans count net rather than total. Some sugar alcohols (erythritol, allulose) also don't impact glucose; others (maltitol) do. When in doubt, count total carbs — it's the safer default.

A 50g net carb target with high fiber is functionally similar to a 30g total carb target. Both will keep most people in ketosis.

## What to eat (high-level)

**Yes:**
- Meat, poultry, fish, eggs
- Healthy fats: olive oil, avocado, butter, ghee, MCT oil
- Above-ground vegetables: leafy greens, broccoli, cauliflower, zucchini, peppers
- Nuts and seeds (in moderation — carbs add up)
- Berries (small portions) — the lowest-carb fruits
- Full-fat dairy: cheese, Greek yogurt, heavy cream
- Avocados

**Limit / avoid:**
- Bread, rice, pasta, grains
- Most fruit (bananas, mangoes, grapes are high-carb)
- Beans, lentils (some keto plans include in moderation)
- Sugar, sweetened anything
- Most "low-fat" packaged foods (often sugar-loaded)

## How Callie handles keto

Callie's tracker can switch into keto mode with one tap — net carbs become the default, the AI coach flags high-carb foods, and the macro targets adjust to your ratio of choice. The voice and photo logging work the same as any other goal mode, so you can say "khichdi with ghee" or "scrambled eggs and bacon" and Callie parses the macros without you doing the math.

## Sources

1. Volek JS, Phinney SD. (2011). *The Art and Science of Low Carbohydrate Living.* Beyond Obesity.
2. Hall KD, et al. (2016). "Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men." *Am J Clin Nutr* 104(2):324-333.
3. Bueno NB, et al. (2013). "Very-low-carbohydrate ketogenic diet v. low-fat diet for long-term weight loss: a meta-analysis of randomised controlled trials." *Br J Nutr* 110(7):1178-1187.
4. Helms ER, et al. (2014). "Evidence-based recommendations for natural bodybuilding contest preparation: nutrition and supplementation." *J Int Soc Sports Nutr* 11:20.
5. Westman EC, et al. (2007). "Low-carbohydrate nutrition and metabolism." *Am J Clin Nutr* 86(2):276-284.
6. Cleveland Clinic. "Ketogenic Diet." https://my.clevelandclinic.org/health/articles/24003-ketogenic-keto-diet

> **Important:** Educational only, not medical advice. Talk to a doctor or registered dietitian before starting keto, especially if you have any of the conditions listed above.

## Blog (articles, how-tos, explainers)

### AI Calorie Apps Underestimate Meals by up to 345 Calories: What the 2026 NIH Study Found

URL: https://www.mycallie.app/en/blog/ai-calorie-app-accuracy-study-2026
Markdown: https://www.mycallie.app/md/en/blog/ai-calorie-app-accuracy-study-2026.md
Published: 2026-08-25 · Updated: 2026-08-25
Tags: AI calorie tracker accuracy, photo calorie tracking, calorie app study, under-logging, food photo recognition

If you have used a photo-based calorie tracker and quietly wondered whether the number was too good to be true, there is now research pointing the same way. **Four widely used photo-based calorie apps underestimated meals by roughly 250 to 345 calories each.**

That is the headline. The detail underneath it is more useful than the headline, and it changes what you should do about it.

## TL;DR

> Researchers at the NIH photographed 102 meals of exactly known composition and logged each one in four popular photo-based apps, without adjusting any portions. Every app came in low, by 250 to 345 calories and about 30 grams of fat. Fat was the problem macro. High-fat and keto meals were worst. The fix is not to abandon photo logging, it is to stop letting the first estimate stand. Review it, add the oil and butter the camera could not see, and correct the portion. That takes seconds and removes most of the error.

## What the researchers actually did

Aaron Hengist and Olivia Charles at the National Institute of Diabetes and Digestive and Kidney Diseases, part of the NIH, built the test around meals whose composition was already known down to the gram. The meals came out of a controlled metabolic kitchen, the kind used for inpatient feeding studies where every ingredient is weighed.

They photographed 102 of those meals and logged each photo in four widely used photo-based calorie apps. Then they compared what the apps reported against the kitchen's own numbers. A later analysis extended the work past 200 meals.

The design matters. Most accuracy claims in this category, including our own [20-meal benchmark](/blog/how-to-track-calories-from-a-photo), compare an app estimate to a kitchen scale reading. A metabolic kitchen is a stricter reference than that.

## The numbers

| Feature | What the study found |
|---|---|
| Average calorie underestimate per meal | 250 to 345 kcal |
| Average fat underestimate per meal | about 30 g |
| Most consistently estimated macro | Carbohydrate |
| Least consistently estimated macro | Fat |
| Meals estimated best | Higher-calorie meals |
| Meals estimated worst | Light meals, and high-fat or keto meals |
| Apps that underestimated | All four |

_Reported at NUTRITION 2026 from 102 metabolic-kitchen meals. Every app erred in the same direction._

Two things stand out.

The first is that all four apps missed in the same direction. That is not four separate bugs, it is a structural property of estimating calories from an image.

The second is that the error scales with how much fat a meal is hiding. Roughly 30 grams of fat is about 270 calories, which accounts for most of the gap on its own. The calorie error is largely a fat error wearing a different label.

## Why a photo cannot see fat

Think about what a camera actually captures. Shape, colour, surface texture, and approximate area. That is enough to recognise chicken, rice, and broccoli. It is not enough to answer the question that determines the calorie count, which is how much fat went into all three.

The specific blind spots:

- **Absorbed oil.** A stir-fry cooked in two tablespoons of oil looks identical to one cooked in a dry pan. That is roughly 240 calories that leave no visual trace.
- **Butter and fats melted into food.** Butter in mashed potato, cream in a soup, ghee in a curry base.
- **Dressing already tossed through.** A salad with the dressing on it looks like a salad.
- **Depth.** Looking down into a bowl tells you the surface area of the food, not how far down it goes.
- **Density.** Two visually identical scoops of rice can differ by a third in weight depending on how tightly packed they are.

Fat is nine calories per gram against four for protein and carbohydrate, so every gram the camera misses costs more than twice as much. It is also the macro most likely to be invisible. Those two facts together explain the whole result.

It also explains the keto finding. A ketogenic meal is deliberately built to carry most of its energy as fat, which is exactly the energy a photograph cannot see. The apps struggled most with the meals where the invisible macro dominated.

The study did not find that AI food recognition fails to identify food. Recognising what is on a plate is close to solved. The gap is in quantifying it, and specifically in quantifying the part of it you cannot photograph.

## The caveat the headlines skipped

Here is the sentence from the research team that most coverage left out, and it is the one that should change how you read the numbers:

> "People using a photo-based tracking app without adjusting the portions or entering amounts of food should take the results with a grain of salt."

The meals were logged from the photo alone. No portion correction, no adding the oil, no editing the entry. That is a deliberate and defensible study design, because it isolates what the model does on its own. It is also the least accurate way anyone actually uses these apps, or should.

So read the result as a ceiling on the error, not a description of your Tuesday. An app you correct before saving is a different instrument from one you let file the estimate unseen.

One more caveat, and it is a real one. This was presented as an abstract at NUTRITION 2026, the American Society for Nutrition's annual meeting in National Harbor, Maryland, in late July 2026. Conference abstracts have not generally been through journal peer review. The direction of the finding is consistent with everything else known about estimating food energy from images. The exact figures should be treated as provisional until the full paper lands.

## What to do about it

A 300-calorie daily error is not a rounding difference. It is larger than the deficit most people are aiming for, which means it can make a genuine deficit look like a plateau and send you hunting for a metabolic explanation that does not exist. We wrote the diagnostic order for that in [calorie deficit but not losing weight](/blog/calorie-deficit-not-losing-weight).

Five corrections, in order of how much they buy you:

1. **Review every estimate before it saves.** This is the whole game. An estimate you never looked at is an estimate you never corrected, and the study is a measurement of exactly that scenario.
2. **Add the fat you know is there.** If you cooked in oil, log the oil. If there is butter in it, log the butter. You do not need to see it to know it went in. This single habit closes most of the gap the study measured.
3. **Correct the portion, not the food.** The recognition is usually right about what the food is. The quantity is where it drifts. Adjust grams or servings and leave the identification alone.
4. **Weigh the few things that swing hardest.** You do not need a scale for a chicken breast. You do need one for oil, nut butter, cheese, granola, and anything else where a small visual difference is a large calorie difference. Full list in [how to track calories accurately](/blog/how-to-track-calories-accurately).
5. **Judge the app on your trend line.** If your logs say 1,800 and your weight is flat for three weeks, the logs are wrong before the formula is. Calibrate against what your body does, not against what one meal estimate says.

Callie does not auto-log a photo. It shows you what it identified, food by food, with portions and macros, and waits for you to confirm or edit. That was a deliberate choice long before this study, and the reason is exactly what the study measured. An uncorrected AI estimate is the failure mode. Making the correction step unavoidable is the fix.

## The honest position on photo logging

It would be easy to read this study as a case against AI calorie tracking. I do not think that holds up, for a reason the study was not designed to measure.

The alternative to an imperfect log is usually no log. Self-reported food intake has been found to be under-reported by 20 to 40 percent across decades of research, and that is people typing entries by hand, carefully, in a study they know they are part of. A 300-calorie error from a photo you actually took is not obviously worse than a 600-calorie error from a manual entry you eyeballed, or than the week you stopped logging entirely because searching a database for every ingredient got old.

The right conclusion is narrower and more useful. Photo logging is a fast first draft, not a final answer. Treat the estimate as a starting point that needs one round of correction, and the speed advantage survives while most of the error does not.

This article is general information, not medical advice. If you have a medical condition, take medication that interacts with diet, are pregnant, or have a history of disordered eating, talk to a qualified healthcare professional before changing how you eat.

## Related reading

- [How to Track Calories From a Photo](/blog/how-to-track-calories-from-a-photo) covers the photo setup that gives any app the best chance, plus our own benchmark.
- [How to Track Calories Accurately](/blog/how-to-track-calories-accurately) is the practical version: what to weigh, what to estimate, and the four hidden-calorie categories.
- [The Complete Guide to AI Calorie Tracking](/guides/ai-calorie-tracking) explains how the underlying computer vision works and where it breaks down.
- [Calorie Deficit but Not Losing Weight?](/blog/calorie-deficit-not-losing-weight) walks the diagnostic order when the numbers and the scale disagree.

## Sources

1. Charles O, Hengist A et al. (2026). "Accuracy of photo-based dietary tracking applications." Presented at NUTRITION 2026, American Society for Nutrition, National Harbor, MD, July 25 to 28, 2026. Press release: https://www.eurekalert.org/news-releases/1136415
2. American Society for Nutrition, NUTRITION 2026 meeting coverage. https://www.sciencedaily.com/releases/2026/07/260726015237.htm
3. HealthDay. "How Accurate Are Photo-Based Calorie Apps? 4 Are Put To The Test." https://www.healthday.com/health-news/nutrition/how-accurate-are-photo-based-calorie-apps-4-are-put-to-the-test
4. Hall KD, Guo J (2017). "Obesity Energetics: Body Weight Regulation and the Effects of Diet Composition." *Gastroenterology*. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5568065/
5. USDA FoodData Central. https://fdc.nal.usda.gov/

---

### GLP-1 and Muscle Loss: The Protein Numbers That Protect Lean Mass (2026)

URL: https://www.mycallie.app/en/blog/glp-1-muscle-loss-protein
Markdown: https://www.mycallie.app/md/en/blog/glp-1-muscle-loss-protein.md
Published: 2026-08-25 · Updated: 2026-08-25
Tags: GLP-1 muscle loss, protein target, semaglutide, tirzepatide, lean mass, GLP-1 nutrition

Weight loss on a GLP-1 works. The question the field has moved on to is **what** you are losing, and the answer is not all fat.

## TL;DR

> Some of the weight you lose on semaglutide or tirzepatide is lean mass. How much varies by study and by measurement method, but the risk is real and it rises when protein intake falls, which is exactly what appetite suppression encourages. The 2025 four-society advisory puts protein at 1.2 to 1.6 g per kg of body weight per day during active loss, and says plainly that protein without resistance training is not enough. The hard part is not the daily number, it is spreading it across four or five small meals when you do not feel like eating any of them.

## The problem in one paragraph

Muscle is metabolically expensive tissue. Your body maintains it in proportion to how much you use it and how much raw material you supply. Cut energy intake sharply, keep training the same, and supply less protein, and the body will let some of that tissue go. GLP-1 medications produce a large, fast energy deficit almost effortlessly, and they do it by removing the hunger signal that would normally push you to eat.

That combination, a large deficit plus a suppressed appetite plus no particular reason to prioritise protein, is the one that costs lean mass.

Reported estimates for how much of GLP-1 weight loss comes from lean tissue vary considerably across studies, and a lot of that spread is methodological. DXA, bioimpedance and MRI do not agree with each other, and "lean mass" includes water and glycogen, both of which drop early in any weight loss and are not muscle. Be sceptical of any single confident percentage, including ones quoted with authority. What is not in dispute is the direction, or the fact that lower protein intake during semaglutide treatment has been associated with greater muscle loss.

When you read that some fraction of GLP-1 weight loss is lean mass, check what was measured and how. Early weight loss on any diet includes glycogen and its associated water, which registers as lean mass on a DXA scan and is not muscle. Studies that measure at three months and studies that measure at twelve produce very different numbers for that reason alone.

## What the 2025 advisory recommends

In May 2025, four professional bodies published a joint clinical advisory, *Nutritional Priorities to Support GLP-1 Therapy for Obesity*: the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association and The Obesity Society.

The protein recommendation is **1.2 to 1.6 g per kg of body weight per day** during active weight reduction.

| Body weight | Daily protein at 1.2 g/kg | Daily protein at 1.6 g/kg |
|---|---|---|
| 60 kg (132 lb) | 72 g | 96 g |
| 70 kg (154 lb) | 84 g | 112 g |
| 80 kg (176 lb) | 96 g | 128 g |
| 90 kg (198 lb) | 108 g | 144 g |
| 100 kg (220 lb) | 120 g | 160 g |
| 110 kg (243 lb) | 132 g | 176 g |

For context, the RDA is 0.8 g/kg. That figure exists to prevent deficiency in a weight-stable adult. It was never meant to describe what protects muscle during rapid weight loss, and treating it as a target here is a mistake.

The advisory also makes a point that gets dropped in most summaries: **protein alone is unlikely to be an adequate muscle-preservation strategy without structured resistance training.** Protein is the material. Training is the signal that the material is still needed. Supply one without the other and you are half-solving the problem.

## Your number

[Interactive protein calculator — use it on the page.]

Use the GLP-1 preset if the calculator offers one. If you are working from body weight in pounds, divide by 2.205 to get kilograms first.

## The part nobody warns you about: distribution

Here is where GLP-1 nutrition stops being ordinary weight-loss nutrition.

Off medication, hitting 130 g of protein is mostly an arithmetic problem. You eat three meals, two of them carry 40 to 50 g, and you are close. On a GLP-1, that plan collapses, because a 50 g protein meal is physically a lot of food and the medication has made a lot of food unappealing or actively uncomfortable.

The workaround is to change the shape of the day rather than the size of the meals. Four or five eating occasions carrying **20 to 40 g each** gets to the same daily total without ever asking you to sit in front of a plate you cannot finish. Muscle protein synthesis responds to per-meal doses in roughly that range anyway, so this is not a compromise, it is arguably the better structure.

| Feature | Three large meals | Four to five smaller occasions |
|---|---|---|
| Protein per eating occasion | 40 to 50 g | 20 to 40 g |
| Realistic on a suppressed appetite | No | Yes |
| Matches per-meal synthesis response | Partially | Yes |
| Survives a bad nausea day | No | Usually |
| Planning effort | Low | Moderate |

_On a GLP-1, the constraint is capacity per sitting, not appetite across the day._

## Hitting the number when you are not hungry

Ranked by how much they help in practice.

**Choose density over volume.** Greek yoghurt over regular. Chicken thigh over chicken soup. Cottage cheese, eggs, fish, lean meat, tofu. When your capacity per sitting is limited, every bite has to carry more. Salad, broth and vegetable-heavy dishes are the wrong shape for this phase, however healthy they are.

**Eat the protein first.** Within any meal, start with the protein while you still have room. What you eat last is what gets abandoned, so it should not be the part that matters.

**Front-load the day.** For many people nausea and fullness build as the day goes on. If that is your pattern, get 30 to 40 g in before lunch rather than hoping dinner will carry it.

**Drink it when you cannot eat it.** A shake, milk, or a fortified soup delivers 25 to 30 g without asking you to chew through it. This is not a lesser option on a GLP-1, it is often the difference between hitting the floor and missing it.

**Keep a default that needs no decision.** A protein source you can produce with no effort on a bad day. The plan that requires cooking is the plan that fails on the day you feel worst.

Persistent nausea, vomiting, an inability to keep food down, rapid unintended loss beyond your target, or new weakness are things to raise with the clinician who prescribed the medication, not to solve with a nutrition article. Dose titration, nutritional adequacy and screening for deficiency are clinical decisions. The advisory also flags micronutrient monitoring as part of ongoing care, which is another reason this belongs in a consultation.

## Why tracking matters more here, not less

There is a reasonable-sounding argument that a GLP-1 removes the need to track. The medication creates the deficit, so why count?

It gets the direction backwards. Tracking on a GLP-1 is not about enforcing a ceiling, it is about defending a floor. Two floors, really: enough protein, and enough total food to stay nourished. Both are easy to breach when nothing is hungry, and neither announces itself. You do not feel yourself missing a protein target the way you feel yourself overeating.

So the practical setup is inverted from ordinary weight loss. Watch protein daily. Watch total intake weekly for a downward drift you did not intend. Let the calorie ceiling look after itself, because it will.

If you want the full eating structure rather than just the protein piece, our [GLP-1 diet plan](/blog/glp-1-diet-plan-ozempic-wegovy-mounjaro) covers food triggers, meal timing, and a sample week. For the general case of protein targets off medication, see the [protein calculator](/blog/protein-calculator).

## The short version

Protein floor of 1.2 to 1.6 g per kg. Four or five occasions of 20 to 40 g rather than three large meals. Resistance training two or three times a week, because the advisory is right that protein without it does not do the job. Track the floor rather than the ceiling.

None of this is exotic. It is just the opposite of the instinct the medication gives you, which is why it has to be deliberate.

This article is general information, not medical advice. GLP-1 medications are prescription drugs with real side effects and contraindications. Decisions about dosing, nutrition targets, supplementation and monitoring belong with your prescribing clinician or a registered dietitian who knows your history.

## Related reading

- [Protein Calculator](/blog/protein-calculator) has the goal presets and a reference table by bodyweight.
- [GLP-1 Diet Plan: What to Eat on Ozempic, Wegovy and Mounjaro](/blog/glp-1-diet-plan-ozempic-wegovy-mounjaro) covers the full eating structure.
- [Calorie Counting vs Macro Tracking](/blog/calorie-counting-vs-macro-tracking) explains why GLP-1 users are one of the few groups where macro tracking genuinely earns its effort.
- [How Long Does It Take to Lose Weight?](/blog/how-long-does-it-take-to-lose-weight) for realistic rates of loss.

## Sources

1. American College of Lifestyle Medicine, American Society for Nutrition, Obesity Medicine Association, The Obesity Society (2025). "Nutritional Priorities to Support GLP-1 Therapy for Obesity." https://nutrition.org/nutritional-priorities-to-support-glp-1-therapy-for-obesity/
2. The Obesity Society (2025). "Nutritional Priorities to Support GLP-1 Therapy for Obesity." https://www.obesity.org/nutritional-priorities-to-support-glp-1-therapy-for-obesity/
3. Morton RW et al. (2018). "A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength." *BJSM*. https://bjsm.bmj.com/content/52/6/376
4. Wilding JPH et al. (2021). "Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1)." *NEJM*. https://www.nejm.org/doi/full/10.1056/NEJMoa2032183
5. Jastreboff AM et al. (2022). "Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1)." *NEJM*. https://www.nejm.org/doi/full/10.1056/NEJMoa2206038
6. USDA FoodData Central. https://fdc.nal.usda.gov/

---

### Why Most People Quit Calorie Tracking Within a Month (And How Not To)

URL: https://www.mycallie.app/en/blog/why-people-quit-calorie-tracking
Markdown: https://www.mycallie.app/md/en/blog/why-people-quit-calorie-tracking.md
Published: 2026-08-25 · Updated: 2026-08-25
Tags: calorie tracking consistency, why people quit tracking, food logging habit, tracking streak, cheat day, adherence

Most people who start calorie tracking are not still doing it a month later. The usual explanation is a lack of discipline. **I think that is the wrong diagnosis, and it leads to the wrong fix.**

## TL;DR

> The thing that determines whether tracking works is whether you are still doing it in three months. Research on technology-based weight loss programmes ties consistency of self-monitoring to outcomes, and finds people disengage fastest from the highest-effort tasks. So the habit ends for structural reasons: it costs too much per meal, one missed day reads as total failure, the effort has no visible payoff, and the target was unrealistic to begin with. Fix those and precision stops mattering much. Ignore them and precision never gets the chance to matter.

## The failure mode that actually matters

There are two ways to get a calorie log wrong. You can log inaccurately, or you can not log at all.

Almost all the advice in this category, including plenty of ours, addresses the first one. Weigh your food. Watch for hidden calories. Do not eyeball the oil. That advice is correct and I would still give it. But it optimises the smaller problem.

Work on self-monitoring in technology-based weight loss programmes has found that consistency of logging predicts weight loss, and that disengagement happens fastest with the tasks that cost the most effort. Dietary logging is the highest-effort task in the set, and it is the first one people drop.

Put those together and the picture is uncomfortable for anyone selling precision. The most consequential thing about your log is whether it exists.

The research supports two claims: consistent self-monitoring is associated with better weight loss outcomes, and lower-burden self-monitoring tasks are sustained longer. It does not directly test "consistent-but-imprecise beats precise-but-sporadic" as a head-to-head. That last step is inference, and I want to be clear it is mine rather than a study's. It follows from the two findings, but nobody has run the trial.

## The four things that end the habit

### 1. It costs too much per meal

Every meal is a small negotiation between the value of logging it and the effort of logging it. Search a database, pick between fourteen entries for "chicken breast," guess a portion, adjust the serving unit. Do that three times a day and the cost side wins eventually. Not dramatically. You just start skipping the annoying meals, then the annoying days.

This is the one worth attacking hardest, because it compounds. Whatever gets logging down to seconds rather than minutes, photo, voice, saved meals, repeated favourites, buys you weeks of habit at the far end.

### 2. One missed day becomes three weeks

The all-or-nothing break is the most destructive pattern in this whole category, and it has almost nothing to do with calories.

You miss Saturday. Now the week is incomplete. An incomplete week feels like a failed week, and a failed week raises the obvious question of whether you should be doing this at all. By Wednesday you have not logged since Friday and the app is a small monument to something you gave up on.

The calorie cost of one unlogged day is trivial. The cost of what you concluded from it is not.

### 3. The effort has no visible return

Logging is work now for a benefit later. Body weight, which is the obvious feedback signal, is noisy enough over days that it frequently tells you the opposite of the truth. Water, sodium, glycogen, and where you are in a menstrual cycle all move the scale more than a day of fat loss does.

So the honest experience of week two is: I did all this work and the number went up. Very few habits survive that without some other visible sign of progress.

### 4. The target was never realistic

A 1,200-calorie target set in a burst of enthusiasm is not a plan, it is a promise made by someone who was not hungry at the time. It works for four days. Then it does not, and because the target is the plan, breaking it feels like the whole thing has collapsed.

If you would not accept the target on a bad week, it is the wrong target. Set the one you would still hit at 70 percent effort.

## Consistency beats precision

Take two people over eight weeks.

One logs every day and is 15 percent low. Their absolute numbers are wrong. Their trend is correct, their awareness of what they eat is intact, and after eight weeks they can compare predicted change against actual change and correct the offset in a single adjustment.

The other logs meticulously for eleven days, misses a weekend, restarts, misses another, and quits in week four. Their eleven days are accurate. Nothing follows from them.

The first person has something to work with. Systematic error is correctable. Missing data is not.

This is also why I would rather you log a restaurant meal as a rough guess than not log it. The guess is wrong. It is still enormously more useful than the gap.

## Make the streak survivable

The all-or-nothing break has a specific fix, and it is to decide in advance which days are exceptions.

An unplanned high day feels like a failure because you did not choose it. The same day, scheduled on Tuesday for the Saturday it happens on, is just part of the week. Same food, same calories, completely different meaning, and the meaning is what determines whether you log on Sunday.

The arithmetic usually supports this too. What moves weight is the weekly average, and one high day inside a week that accounts for it is absorbable.

[Interactive macro calculator with cheat-day adjuster — use it on the page.]

You can mark a day as a cheat day in advance and your streak stays intact. That is not a loophole, it is the point. A habit that breaks the first time life happens is not a habit worth building, and two months of consistency should not be erased by one birthday dinner.

## Make the progress visible

The second structural fix is giving the effort a payoff you can see on a day when the scale is unhelpful.

A weight number is a poor daily feedback signal for the reasons above. A record of the behaviour itself is a good one, because the behaviour is the thing you actually control. Did you log today, yes or no. That answer is never ambiguous and never lies to you the way a morning weigh-in does.

Laid out as a grid across weeks, it also shows you something a number cannot: your own pattern. Most people find their gaps cluster somewhere specific. Weekends, or travel, or the days after a bad night's sleep. That is actionable in a way "I am bad at this" never is.

Callie's streak dashboard is a contribution-graph style grid for exactly this. Filled squares for days you logged, gaps for days you did not, longest streak counted. It is deliberately not a notification that scolds you.

## If you have already stopped

Most people reading this are not starting, they are restarting. Three rules.

**Start today, not Monday.** Waiting for a clean boundary is the same instinct that made one missed day feel disqualifying. There is no clean boundary.

**Do not back-fill.** Reconstructing last week from memory produces bad data and makes restarting feel like homework. The old data is gone. That is fine.

**Do not run a penance deficit.** Cutting extra calories to make up for the time off is the reflex, and it is the reliable way to end up back here in three weeks. Return to the same target you had.

For some people, tracking of any kind feeds an unhealthy relationship with food. If logging has tipped into preoccupation for you, if a missed day produces real distress, or if you find yourself compensating with restriction, that is a reason to stop and talk to a clinician rather than a reason to build a better streak. This is not a rare edge case in this category.

## The short version

Get the cost per meal down. Plan the exceptions so a break in the pattern is part of the pattern. Give yourself something visible in return for the work. Set a target you would accept on a bad week.

Do those and you will probably log imprecisely for a long time, which is the version that works. The alternative is logging precisely until you stop.

This article is general information, not medical advice. If you have a medical condition, take medication that interacts with diet, are pregnant, or have a history of disordered eating, talk to a qualified healthcare professional before changing how you eat.

## Related reading

- [Macro Calculator with Cheat Day Adjuster](/blog/macro-calculator-with-cheat-day) does the weekly-average maths around a planned high day.
- [How to Lose Weight With a Calorie Tracker](/guides/how-to-lose-weight-with-a-calorie-tracker) is the full pillar on deficits, timelines and the anti-obsession approach.
- [Why Have I Stopped Losing Weight?](/blog/weight-loss-plateau) separates a real plateau from the noise that makes people quit.
- [How to Track Calories Accurately](/blog/how-to-track-calories-accurately) for when you are ready to tighten the numbers up.

## Sources

1. Painter SL, Turner D et al. (2022). "Consistency With and Disengagement From Self-monitoring of Weight, Dietary Intake, and Physical Activity in a Technology-Based Weight Loss Program: Exploratory Study." *JMIR Formative Research*. https://formative.jmir.org/2022/2/e33603
2. Chin SO et al. (2016). "Successful weight reduction and maintenance by using a smartphone application in those with overweight and obesity." *Scientific Reports*. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5069507/
3. Patel ML et al. (2021). "Assessing the Contribution of Self-Monitoring Through a Commercial Weight Loss App: Mediation and Predictive Modeling Study." *JMIR mHealth and uHealth*. https://mhealth.jmir.org/2021/7/e18741
4. Burke LE, Wang J, Sevick MA (2011). "Self-monitoring in weight loss: a systematic review of the literature." *Journal of the American Dietetic Association*. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3268700/

---

### Calorie Counting vs Macro Tracking: Which Should You Actually Do? (2026)

URL: https://www.mycallie.app/en/blog/calorie-counting-vs-macro-tracking
Markdown: https://www.mycallie.app/md/en/blog/calorie-counting-vs-macro-tracking.md
Published: 2026-08-04 · Updated: 2026-08-04
Tags: calorie counting vs macro tracking, macro tracking, calorie counting, protein target, IIFYM

The short version: **calories decide whether you lose weight. Macros decide what you lose and how you feel while doing it.**

Almost every argument about these two methods happens because people are answering different questions. "Will I lose weight?" and "will I lose fat rather than muscle, and will I be miserable?" have different answers, and only the second one needs macros.

## TL;DR

> Track calories plus a protein target. That's the highest return per unit of effort for most people, and it's where you should start. Add full carbohydrate and fat tracking only if you're doing body recomposition, following keto or another therapeutic diet, using a GLP-1 medication, or programming an athletic cut or bulk. And before optimizing either method, fix your logging accuracy — under-logging by 20–40% is far more common than picking the wrong framework.

## What each method actually does

| Feature | Calorie counting | Macro tracking |
|---|---|---|
| Drives weight loss | Yes | Yes |
| Protects lean mass in a deficit | Partially | Yes |
| Manages hunger deliberately | No | Yes |
| Effort per meal | Low | Moderate |
| Learning curve | Hours | Weeks |
| Works for keto / low carb | No | Yes |
| Useful for athletic programming | Limited | Yes |
| Risk of over-focus on numbers | Moderate | Higher |
| Needed to hit a goal weight | Yes | No |

_Both methods require a calorie total; macro tracking simply adds structure to how those calories are composed._

Worth saying explicitly, because it confuses a lot of people: **macro tracking includes calorie counting.** Protein and carbohydrate are 4 calories per gram, fat is 9. If you hit your macros, you've hit a calorie total by definition. Macro tracking isn't an alternative to counting calories — it's counting calories with additional constraints.

## What the research says about macronutrient splits

This is where the debate mostly gets settled. When trials control total calories and protein, the specific carbohydrate-to-fat split has little effect on weight loss.

- **POUNDS LOST** (Sacks et al., 2009) randomized 811 adults to four diets varying widely in fat, protein, and carbohydrate. After two years, weight loss was similar across all four.
- **DIETFITS** (Gardner et al., 2018) randomized 609 adults to healthy low-fat or healthy low-carb for 12 months. Mean weight change differed by less than 2 kg between groups, with enormous individual variation inside each group.
- A **JAMA meta-analysis** of named diets (Johnston et al., 2014) covering 48 trials found differences between branded diets were small and largely disappeared at 12 months.

The practical read: the split you can adhere to beats the theoretically optimal split you abandon in three weeks.

Protein is the exception, and it's a real one. Higher protein intake during a deficit consistently preserves more lean mass and increases satiety per calorie. Morton et al. (2018), a meta-analysis of 49 studies, found benefits to lean mass gain up to roughly 1.6 g per kg of bodyweight per day in resistance-trained individuals. Reviews of aggressive cuts in lean athletes support going higher still.

Set a calorie target. Set a protein floor. Ignore the carb-to-fat split unless you have a specific reason to care. This is what most successful trackers converge on after a year, and you can start there instead.

## Your protein target

Since protein is the macro that actually earns its tracking effort, start here.

[Interactive protein calculator — use it on the page.]

## When full macro tracking is worth it

Four cases where the extra effort pays for itself.

### 1. Body recomposition

Losing fat and gaining muscle simultaneously is the one goal where macros are close to mandatory. It requires a small deficit, a high protein intake, and enough carbohydrate to train hard. Calories alone won't get you there — you can hit a deficit on 60 g of protein a day and lose a meaningful amount of muscle doing it.

### 2. Keto or another therapeutic diet

Ketosis depends on keeping carbohydrate under a threshold, usually 20–50 g per day. That's a macro constraint, not a calorie one. You cannot run keto by counting calories. Same logic applies to low-FODMAP, medically prescribed low-protein diets, and diabetes management where carbohydrate distribution across the day matters.

Our [keto macro calculator](/guides/keto-macro-calculator) handles the ratios.

### 3. GLP-1 medication users

If you're on semaglutide or tirzepatide, appetite suppression makes the calorie deficit almost automatic — and that's precisely the problem. Total intake can fall well below what's needed to preserve lean mass, and studies of GLP-1 weight loss show a meaningful fraction coming from lean tissue.

Here, protein tracking isn't optional. Hitting a protein floor when you don't feel hungry requires deliberate planning. See our [GLP-1 diet plan](/blog/glp-1-diet-plan-ozempic-wegovy-mounjaro) for the specifics.

### 4. Athletes in a structured cut or bulk

If training performance is an outcome you care about, carbohydrate timing and total matter. A powerlifter in a cut and a marathoner in a build have different carb needs, and calorie counting alone doesn't express that.

## Effort versus benefit, by goal

| Your goal | Recommended approach | Why |
|---|---|---|
| Lose 10–30 lb, no athletic goal | Calories + protein floor | Full macro tracking adds effort without changing the outcome |
| Body recomposition | Full macros | Protein and carbs both need to be deliberate |
| Keto / low carb | Full macros | Carb ceiling is the whole mechanism |
| On a GLP-1 medication | Calories + strict protein floor | Appetite suppression makes under-eating protein the main risk |
| Athletic cut or bulk | Full macros | Carb total and timing affect training quality |
| Maintenance after weight loss | Calories, loosely | Consistency matters more than precision here |
| History of disordered eating | Neither — talk to a clinician first | Numeric tracking can be actively harmful for some people |

For a subset of people, detailed tracking of any kind reinforces an unhealthy relationship with food. If tracking has previously tipped into obsession for you, or you notice it doing so now, that's a reason to stop and talk to a clinician — not a reason to track harder. This is not a niche concern in this category.

## The thing that matters more than either method

Accuracy of your logs, by a wide margin.

Research on self-reported food intake consistently finds people under-report by 20–40%. A 30% under-log on a 2,000-calorie target is 600 calories — larger than the entire deficit most people are aiming for, and larger than any conceivable difference between calorie counting and macro tracking.

The usual sources: cooking oils and butter, drinks, sauces and dressings, bites while cooking, restaurant portions being bigger than the database entry, and weekends logged with less care than weekdays.

If you're deciding between these two methods, fix logging accuracy first. We wrote about the specifics in [how to track calories accurately](/blog/how-to-track-calories-accurately), and if your deficit isn't producing results, [calorie deficit but not losing weight](/blog/calorie-deficit-not-losing-weight) works through the diagnostic order.

## Planning around a cheat day

The most common failure point for both methods is a single high day that feels like it ruined the week. Usually it didn't — what matters is the weekly average, and one planned high day is absorbable if the rest of the week accounts for it.

[Interactive macro calculator with cheat-day adjuster — use it on the page.]

## How to start, concretely

1. **Get your maintenance calories.** Use the [TDEE calculator](/blog/maintenance-calories-tdee-calculator) or the fuller [calorie calculator guide](/guides/calorie-calculator).
2. **Subtract 15–25%** for fat loss. Aggressive deficits mostly buy you more muscle loss and worse adherence.
3. **Set a protein floor** from the calculator above. Treat it as a minimum, not a target to hit exactly.
4. **Log for two weeks without changing anything else.** You're collecting a baseline, not optimizing yet.
5. **Compare actual weight change to predicted.** If they disagree, the logs are wrong before the formula is.
6. **Add carb and fat targets only if** one of the four cases above applies to you.

## Verdict

Calorie counting is the floor and it's sufficient for most weight-loss goals. Macro tracking is a specialization that pays off for recomposition, therapeutic diets, GLP-1 users, and athletes.

The version almost everyone should run: **calories plus protein.** It's most of the benefit of macro tracking with a fraction of the friction, and friction is what determines whether you're still doing this in six months.

This article is general information, not medical advice. If you have a medical condition, take medication that interacts with diet, are pregnant, or have a history of disordered eating, talk to a qualified healthcare professional before changing how you eat.

## Sources

1. Sacks FM et al. (2009). "Comparison of Weight-Loss Diets with Different Compositions of Fat, Protein, and Carbohydrates." *NEJM*. https://www.nejm.org/doi/full/10.1056/NEJMoa0804748
2. Gardner CD et al. (2018). "Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss (DIETFITS)." *JAMA*. https://jamanetwork.com/journals/jama/fullarticle/2673150
3. Johnston BC et al. (2014). "Comparison of Weight Loss Among Named Diet Programs." *JAMA*. https://jamanetwork.com/journals/jama/fullarticle/1900510
4. Morton RW et al. (2018). "A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength." *BJSM*. https://bjsm.bmj.com/content/52/6/376
5. Helms ER et al. (2014). "Evidence-based recommendations for natural bodybuilding contest preparation: nutrition and supplementation." *JISSN*. https://jissn.biomedcentral.com/articles/10.1186/1550-2783-11-20
6. Hall KD, Guo J (2017). "Obesity Energetics: Body Weight Regulation and the Effects of Diet Composition." *Gastroenterology*. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5568065/
7. USDA FoodData Central. https://fdc.nal.usda.gov/

---

### MyFitnessPal vs Cal AI (2026): Same Owner, Two Apps — Which Should You Use?

URL: https://www.mycallie.app/en/blog/myfitnesspal-vs-cal-ai
Markdown: https://www.mycallie.app/md/en/blog/myfitnesspal-vs-cal-ai.md
Published: 2026-08-04 · Updated: 2026-08-04
Tags: MyFitnessPal vs Cal AI, Cal AI acquisition, MyFitnessPal, Cal AI, AI calorie tracker

For most of 2024 and 2025 these two apps were the clearest rivals in calorie tracking: the decade-old database incumbent versus the viral photo-first newcomer. That framing is out of date. They now have the same owner.

This comparison is written by the team behind Callie, a competing app. We've kept the head-to-head itself neutral and sourced, and flagged clearly where we talk about our own product.

## TL;DR

> MyFitnessPal owns Cal AI, but they are still two separate apps with separate subscriptions and different jobs. Use Cal AI if you want to log by photo in a few seconds and don't want to think about a database. Use MyFitnessPal if you log branded packaged foods, want the largest food database and a web interface, or need broad wearable integrations. Neither includes AI logging in its free tier.

## What actually happened

The deal closed in December 2025 after roughly a year of negotiation and was announced publicly on March 2, 2026. Financial terms were not disclosed.

The timeline that matters:

| When | What happened |
|---|---|
| Early 2025 | Acquisition talks begin |
| December 2025 | Deal closes |
| March 2, 2026 | Acquisition announced publicly |
| April 2026 | Apple removes Cal AI from the App Store over billing violations, then restores it after fixes |

Cal AI was built by Zach Yadegari and Henry Langmack, who started it while still in high school. By the time of the acquisition it had passed 15 million downloads in under two years and roughly $30 million in annual revenue. Its entire seven-person team, including Yadegari as CEO, was retained.

The two things MyFitnessPal has publicly committed to are worth noting, because they answer the question most existing users have:

1. **Cal AI keeps operating as a standalone app.** There is no plan to merge it into the MyFitnessPal app or migrate users across.
2. **Cal AI now runs on MyFitnessPal's nutrition database** — 20 million+ foods, 68,500 brands, and 380+ restaurant chains.

That second point is the substantive product change. Cal AI's weakest link was always what happened *after* the vision model identified your food: it needed nutrition data to attach numbers to. It now draws on the largest database in the category.

### The April 2026 App Store removal

Worth understanding if you're evaluating Cal AI, because it was about billing rather than the tracking itself.

Apple removed the app in April 2026 and told TechCrunch it had found several violations:

- The app bypassed Apple's required in-app purchase flow, using a Stripe-based checkout instead. External payments are permitted after the Epic ruling, but Cal AI is not a "reader app," so it was required to offer in-app purchase alongside.
- Weekly pricing was displayed more prominently than the amount actually charged.
- The free-trial toggle did not make automatic renewal clear.
- Users who declined were re-prompted with a second, different subscription offer.

Apple was explicit that the problem was the deceptive billing design, not the use of external payments. Cal AI was restored after the developer made fixes, and subsequently ranked #4 on the App Store's Health and Fitness chart. Neither MyFitnessPal nor Cal AI commented publicly.

## At-a-glance comparison

| Feature | Cal AI | MyFitnessPal |
|---|---|---|
| Primary logging model | Photo first | Search / barcode first |
| AI photo logging | Core feature | Premium (meal scan) |
| Voice logging | No | Premium |
| Barcode scanner | Yes | Free for single items |
| Food database | MyFitnessPal database (since 2026) | 20.5M+ foods |
| Restaurant chain coverage | Via MyFitnessPal database | 380+ chains |
| Web / desktop access | No | Yes |
| Wearable + app integrations | Limited | Extensive |
| AI logging in free tier | No | No |
| App Store rating | 4.8 (349K ratings) | 4.7 (2.3M ratings) |
| Minimum iOS | 18.0 | 17.0 |
| Owned by | MyFitnessPal | Francisco Partners |

_Compiled from both apps' App Store listings and public reporting, August 2026. Feature splits and pricing change frequently — confirm in-store before subscribing._

## Pricing

Both apps are free to download and both gate their AI logging behind a subscription. Exact tier mapping shifts with promotions and region, so the honest presentation is the range each App Store listing shows.

| Feature | Cal AI | MyFitnessPal |
|---|---|---|
| Free to download | Yes | Yes |
| In-app purchases listed | $0.99 – $29.99 | $9.99 – $79.99 |
| Monthly options | Weekly and monthly tiers | $9.99 / $19.99 |
| Annual options | See listing | $49.99 / $79.99 |
| Tiers | Single subscription | Premium and Premium+ |
| Ads in free tier | No | Yes |
| One subscription covers both apps | No | No |

_In-app purchase ranges as listed on the US App Store, August 2026. These lists include promotional and regional prices, so treat them as a range rather than a tier map. Confirm the current price in-app before subscribing._

In May 2026 MyFitnessPal also moved three more features into Premium: scan-a-meal (photographing a whole plate), recipe URL import, and macro-by-meal goals. Single-item barcode scanning, manual entry, and the food database itself remain free.

## Head-to-head on each dimension

### Speed

Cal AI wins clearly, and it's the reason the app grew as fast as it did. Photograph the plate, get numbers. There's no search box, no scrolling past nine near-identical crowdsourced entries for "grilled chicken breast."

MyFitnessPal is faster than Cal AI for exactly one case: a packaged food with a barcode. Scanning is both quicker and more accurate than any photo estimate.

### Accuracy

These fail differently, which is more useful to understand than a single ranking.

- **Cal AI** estimates portion and content from an image. Category-wide, photo estimation lands within roughly 10–25% of a weighed reference, and degrades on layered or mixed dishes — casseroles, curries, anything where the calories are hidden under the surface. Oils and sauces are the usual invisible cost.
- **MyFitnessPal** is exactly as accurate as the entry you select. A verified barcode scan is near-exact. A crowdsourced entry for someone else's homemade lasagna can be off by a wide margin, and the database's duplicate-entry problem means picking well is a learned skill.

Neither app is a measuring instrument. Both are useful for spotting trends across weeks.

### Database

This used to be MyFitnessPal's decisive advantage over Cal AI. After the acquisition it's a shared asset — 20 million+ foods, 68,500 brands, 380+ restaurant chains, now behind both apps. If database coverage was your reason to prefer MyFitnessPal over Cal AI, that reason is substantially weaker in 2026 than it was in 2025.

### Where you can log

MyFitnessPal has a web interface. Cal AI is mobile-only. If you like reviewing or bulk-editing a week of logs on a laptop, that's a real difference.

### Integrations

MyFitnessPal has a decade of integration work behind it — Garmin, Withings, Strava, Apple Health, Google Fit and a long tail beyond. Cal AI's integration surface is much smaller. Check both against whatever wearable you actually own.

### Ads

MyFitnessPal's free tier shows ads. Cal AI's does not.

## What the acquisition means for you

Three honest implications.

**One company now owns both the largest food database and the most-downloaded photo tracker.** That's good for Cal AI users, who inherited real nutrition data. It also means less competitive pressure in a category where the main direction of travel has been features moving from free to paid.

**Your data doesn't move, and neither does your subscription.** Separate apps, separate billing. Nothing about the acquisition changes what you're paying for.

**"Which of these two is independent?" is no longer a question with an answer.** MyFitnessPal has been owned by the private equity firm Francisco Partners since 2020, and Cal AI is now inside that structure. If you were choosing between an incumbent and a scrappy startup, that choice no longer exists in this pairing.

## Which should you use?

**Choose Cal AI if** you log mostly home-cooked or restaurant meals, you want the fastest possible photo flow, you're on iOS 18 or later, and you don't need web access or deep integrations.

**Choose MyFitnessPal if** you log mostly branded packaged foods, you want the free single-item barcode scanner, you want a web interface, or you depend on a specific third-party integration.

**Choose neither if** free-tier AI logging is your requirement. Both apps paywall it.

## A third option

Since that last point is where we live, we'll be direct about it: Callie is our app, and its AI logging is in the free tier.

The relevant differences, stated plainly:

- **Photo, voice, and text logging are all free**, rather than photo-only or Premium-only.
- **Review before logging.** Callie shows the detected foods, portions, calories and macros on a review screen and lets you edit before anything is saved. Given that photo estimation is 10–25% off across the whole category, catching the error before it enters your data matters more than the model's raw accuracy. Cal AI's flow is more automatic.
- **A localized UI** in English, French, Spanish, German and Arabic, with an AI coach that chats in any language. Both apps here are functionally English-first.
- **Habit features neither has**: a GitHub-style streak dashboard, scheduled cheat days that preserve your streak, and a daily diet plan matched to your BMI.

Where we don't win: MyFitnessPal's database is larger than ours, it has a web app and we don't, and its integration list is longer.

Longer head-to-heads if you want them: [Callie vs Cal AI](/compare/callie-vs-cal-ai) and [Callie vs MyFitnessPal](/compare/callie-vs-myfitnesspal). Broader field: [Best AI Calorie Tracker Apps of 2026](/blog/best-ai-calorie-tracker-apps-2026).

## Sources

1. TechCrunch (March 2, 2026). "MyFitnessPal has acquired Cal AI, the viral calorie app built by teens." https://techcrunch.com/2026/03/02/myfitnesspal-has-acquired-cal-ai-the-viral-calorie-app-built-by-teens/
2. TechCrunch (April 21, 2026). "Apple's Cal AI crackdown signals it's still policing the App Store." https://techcrunch.com/2026/04/21/apples-cal-ai-crackdown-signals-its-still-policing-the-app-store/
3. MacRumors (April 21, 2026). "Apple Pulled Cal AI for Deceptive Billing Design, Not External Payments." https://www.macrumors.com/2026/04/21/apple-cal-ai-app-store-removal/
4. MyFitnessPal App Store listing (accessed August 2026). https://apps.apple.com/us/app/myfitnesspal-calorie-counter/id341232718
5. Cal AI App Store listing (accessed August 2026). https://apps.apple.com/us/app/cal-ai-calorie-tracker/id6480417616
6. USDA FoodData Central. https://fdc.nal.usda.gov/

---

### MyFitnessPal vs Lose It! (2026): What You Actually Get for Free

URL: https://www.mycallie.app/en/blog/myfitnesspal-vs-lose-it
Markdown: https://www.mycallie.app/md/en/blog/myfitnesspal-vs-lose-it.md
Published: 2026-08-04 · Updated: 2026-08-04
Tags: MyFitnessPal vs Lose It, MyFitnessPal alternative, Lose It, MyFitnessPal, free calorie tracker

Most comparisons of these two apps were written when their free tiers were much more generous. Both have tightened since, and the specific things that got gated are the things that make logging fast.

So this comparison starts with the free tier, because that's where the 2026 decision actually gets made.

We build a competing app, Callie. The comparison below is based on both companies' current App Store listings, and we've kept our own product to a clearly-labeled section at the end.

## TL;DR

> MyFitnessPal has the better free tier — single-item barcode scanning and a 20.5M+ food database stay free, at the cost of ads. Lose It!'s free tier is close to plain calorie logging, since Snap It, AI voice logging, the barcode scanner and macros are all Premium. Lose It! is the cheaper paid app and the only one of the two with a lifetime option. Neither includes AI logging for free.

## What's actually free in each app

This is the table to read first.

| Feature | MyFitnessPal | Lose It! |
|---|---|---|
| Manual food logging | Yes | Yes |
| Food database access | Free (20.5M+ foods) | Free |
| Barcode scanner | Free (single items) | Premium |
| Scan a whole meal / plate | Premium | Premium (Snap It) |
| AI photo logging | Premium | Premium |
| AI voice logging | Premium | Premium |
| Macro tracking | Free (custom goals Premium) | Premium |
| Macro-by-meal goals | Premium (since May 2026) | Premium |
| Recipe URL import | Premium (since May 2026) | Premium |
| Intermittent fasting tools | Limited | Premium |
| Meal planning | Premium+ | Premium |
| Wearable / fitness app sync | Free (core), broad support | Premium |
| Web / desktop access | Yes | Yes |
| Ad-free | Premium | Premium |

_Compiled from each app's US App Store listing, August 2026. Both companies restructure feature splits regularly, and rollouts sometimes differ for existing vs new users — confirm in-app._

The pattern is the same in both apps: **anything that removes typing is a paid feature.** Free users in 2026 are manual-entry users.

MyFitnessPal's free tier comes out ahead for one concrete reason — you can still scan a packaged item's barcode without paying, and barcode scanning is the single most accurate logging method either app offers. Lose It! moved its scanner (Scan It) into Premium.

A lot of currently-ranking articles still describe Lose It! as having free barcode scanning and free macro tracking, and describe MyFitnessPal's free tier as capping you at a handful of foods per day. Those claims don't match either app's current App Store listing. Feature splits in this category change every few months; check the listing rather than a blog post, including this one.

## Pricing

| Feature | MyFitnessPal | Lose It! |
|---|---|---|
| Free to download | Yes | Yes |
| In-app purchase range | $9.99 – $79.99 | $9.99 – $79.99 |
| Monthly | $9.99 / $19.99 | $9.99 |
| Annual | $49.99 / $79.99 | $39.99 commonly listed |
| Lifetime purchase | No | $49.99 / $59.99 |
| Tiers | Premium, Premium+ | Premium |
| Free trial | See listing | 7-day trial commonly offered |

_In-app purchase ranges as listed on the US App Store, August 2026. These lists include promotional and regional prices, so read them as ranges rather than a tier map._

Lose It! is the cheaper paid app, and the lifetime option is genuinely unusual in a category that has moved almost entirely to recurring subscriptions. If you know you want to track for years, the arithmetic favors it strongly.

MyFitnessPal's Premium pricing has roughly doubled from the $9.99/month of its Under Armour era. It has been owned by the private equity firm Francisco Partners since 2020.

## Head-to-head on each dimension

### Food database

MyFitnessPal wins, and it's still the app's strongest asset: 20.5 million+ foods, 68,500 brands, 380+ restaurant chains. The cost of a crowdsourced database that size is duplicate and inconsistent entries, so learning to spot verified entries is part of using it well.

Lose It!'s database is large and generally cleaner, but smaller on obscure regional and packaged products.

### Speed

On paid plans, roughly comparable — both offer photo and voice logging.

On free plans, both are slow in the same way: search, scroll, pick an entry, set a portion. Expect around 45 seconds per meal. MyFitnessPal is meaningfully faster for packaged food because of the free scanner.

### Interface

Lose It! is the friendlier app for beginners — cleaner, less dense, fewer upsell surfaces. MyFitnessPal packs more in and pushes Premium harder. Lose It! holds a 4.8 rating from 770K ratings; MyFitnessPal a 4.7 from 2.3 million.

### Integrations

MyFitnessPal has the longer list and keeps core syncing on the free tier. Lose It! places fitness app syncs in Premium. If you own a Garmin, Withings or Strava-linked device, check both lists before paying.

### Platform requirements

Lose It! requires iOS 18.0 or later; MyFitnessPal supports iOS 17.0. Lose It! also ships iPad, Mac, Apple Watch and Vision Pro builds.

### Photo logging quality

Both use AI photo estimation, and both are subject to the same category limits — roughly 10–25% error against a weighed reference, worst on layered and mixed dishes where calories hide below the surface. MyFitnessPal's scan-a-meal became Premium-only in May 2026. Neither app makes it easy to review and correct the estimate before it lands in your daily total, which is where most silent under-logging comes from.

## Which should you pick?

**Choose MyFitnessPal if** you're staying on the free tier, you log a lot of packaged food, you want the largest database, or you need a specific integration. Accept the ads.

**Choose Lose It! if** you're going to pay, you want the lowest cost or a lifetime purchase, and you prefer a simpler interface. It's also the better first tracker for someone who has never counted calories.

**Look elsewhere if** you want AI logging without a subscription, or you want to log in a language other than English. Both apps are functionally English-first, and both paywall the fast logging paths.

## If you're leaving MyFitnessPal anyway

This is the part where we talk about our own app, so weigh it accordingly.

Callie's AI logging is in the free tier — photo, voice, and natural-language text. That's the specific gap the two apps above share. Beyond that:

- **You review before anything is logged.** Callie shows detected foods, portions, calories and macros on a review screen; you edit, then confirm. Since photo estimation is 10–25% off across every app in the category, correcting the estimate matters more than which model produced it.
- **Localized UI** in English, French, Spanish, German and Arabic, plus an AI coach that chats in any language.
- **Habit features** neither app has: a GitHub-style streak dashboard, scheduled cheat days that keep your streak intact, and a daily diet plan matched to your BMI, weight and allergies.
- **No ads** in the free tier.

Where we don't win: MyFitnessPal's database is bigger than ours, Lose It!'s lifetime purchase has no equivalent from us, and both have longer integration lists.

Full head-to-heads: [Callie vs Lose It!](/compare/callie-vs-lose-it) and [Callie vs MyFitnessPal](/compare/callie-vs-myfitnesspal). Also relevant, since the two apps in this article now share an owner with a third: [MyFitnessPal vs Cal AI](/blog/myfitnesspal-vs-cal-ai).

## Sources

1. MyFitnessPal App Store listing (accessed August 2026). https://apps.apple.com/us/app/myfitnesspal-calorie-counter/id341232718
2. Lose It! App Store listing (accessed August 2026). https://apps.apple.com/us/app/lose-it-calorie-counter/id297368629
3. MyFitnessPal Help Center — feature and plan documentation. https://support.myfitnesspal.com/
4. TechCrunch (March 2, 2026). "MyFitnessPal has acquired Cal AI, the viral calorie app built by teens." https://techcrunch.com/2026/03/02/myfitnesspal-has-acquired-cal-ai-the-viral-calorie-app-built-by-teens/
5. USDA FoodData Central. https://fdc.nal.usda.gov/

---

### How Long Does It Take to Lose Weight? (Realistic Timelines, 2026)

URL: https://www.mycallie.app/en/blog/how-long-does-it-take-to-lose-weight
Markdown: https://www.mycallie.app/md/en/blog/how-long-does-it-take-to-lose-weight.md
Published: 2026-06-30 · Updated: 2026-06-30
Tags: how long does it take to lose weight, how long to lose 20 pounds, weight loss timeline, safe rate of weight loss

"How long until I see results?" is the first thing everyone wants to know. Here's the honest math, realistic timelines for common goals, and why the first weeks lie to you.

## The 30-second answer

**A safe, sustainable rate is 0.5–1% of bodyweight per week — about 1–2 lb (0.5–1 kg).** At that pace:

- **10 lb / 4.5 kg:** roughly 5–10 weeks
- **20 lb / 9 kg:** roughly 10–20 weeks (2.5–5 months)
- **50 lb / 23 kg:** roughly 6–12 months

The first week or two often drops faster — but that's water, not fat. Plan around the steady rate. Use the calculator to find your deficit, then the tables below for your timeline.

[Interactive calorie deficit calculator — use it on the page.]

## Realistic timelines by goal

Based on losing **1 lb/week (slower, very sustainable)** vs **2 lb/week (faster, for those with more to lose):**

| Goal | At 1 lb/week | At 2 lb/week |
|---|---|---|
| 5 lb / 2.3 kg | ~5 weeks | ~2–3 weeks |
| 10 lb / 4.5 kg | ~10 weeks | ~5 weeks |
| 20 lb / 9 kg | ~20 weeks | ~10 weeks |
| 30 lb / 14 kg | ~30 weeks | ~15 weeks |
| 50 lb / 23 kg | ~50 weeks | ~25 weeks |

**Important:** the 2 lb/week column is realistic mainly for people with more weight to lose. The closer you get to a lean bodyweight, the harder fast loss becomes — and the more it costs you in muscle. Most people should anchor to the 1 lb/week column.

## Why the first weeks are faster (and misleading)

If you drop 4 lb in week one, don't expect that every week. Here's what's happening:

- **Glycogen and water.** Your body stores carbohydrate as glycogen, and each gram of glycogen holds ~3 g of water. A calorie deficit burns through glycogen fast, releasing that water — so the scale plunges, but it's mostly fluid.
- **Reduced food volume.** Less food in your digestive tract weighs less. Again, not fat.

By weeks 2–3 this settles, and you start seeing your *true* fat-loss rate. This is exactly why people get discouraged at week three: nothing's wrong — the water whoosh just ended and reality set in.

## Why bigger bodies lose faster — then everyone slows down

Two people eating the same deficit won't lose at the same rate:

- A **heavier person** has a higher maintenance, so a given deficit is a smaller *percentage* of intake and the absolute fat loss is larger. They often lose 2+ lb/week early.
- As **anyone gets lighter**, maintenance falls, the deficit shrinks, and the rate slows.

This deceleration is built into every weight-loss journey — it's not failure, it's physics. (When it stalls completely, that's a [plateau, with specific fixes](/blog/weight-loss-plateau).)

## Why faster is not better

It's tempting to crash-diet for speed. It backfires:

- **Muscle loss.** Aggressive deficits without enough protein burn muscle along with fat, lowering your metabolism and leaving you "skinny-fat."
- **Energy and adherence crash.** Very low intake tanks mood, sleep, and willpower — the things you need to keep going.
- **Rebound.** Rapid loss correlates with rapid regain. The weight you lose in 3 weeks of misery tends to return.

A moderate deficit held for 6 months beats an aggressive one abandoned in 3 weeks — every time. The math only works if you're still doing it next month. Pick a pace you could imagine sustaining, then prove it for two weeks.

## What actually speeds it up (safely)

You can support the steady rate without crash-dieting:

1. **Track accurately.** Under-logging silently shrinks your deficit and slows everything. [Fix your tracking accuracy.](/blog/how-to-track-calories-accurately)
2. **Hit protein** (1.6–2.2 g/kg). Preserves muscle so your metabolism stays up and you lose fat, not lean mass.
3. **Add NEAT.** A few thousand more daily steps widens your deficit without formal exercise.
4. **Sleep 7–9 hours.** Poor sleep slows fat loss and spikes hunger.
5. **Stay consistent.** Two solid weeks beat one perfect day and one blowout.

## How Callie keeps you on the timeline

Tell Callie your goal and target date, and it sets a deficit matched to a *safe* rate — then re-tunes every two weeks as your weight (and maintenance) change, so your timeline stays realistic instead of stalling silently. The **streak dashboard** turns the long middle stretch — where motivation fades — into daily momentum, and **streak-preserving cheat days** let you have a big meal without derailing the plan or your streak. Accurate photo and voice logging keeps the deficit honest so the timeline you're promised is the one you actually get.

## Related reading

- [How Many Calories Do I Burn a Day? (TDEE)](/blog/maintenance-calories-tdee-calculator) — the number your timeline is built on.
- [How Many Calories Should I Eat to Lose Weight?](/blog/how-many-calories-to-lose-weight) — set the deficit.
- [Why Have I Stopped Losing Weight?](/blog/weight-loss-plateau) — when the timeline stalls.
- [How to Track Calories Accurately](/blog/how-to-track-calories-accurately) — keep the pace honest.

## Sources

1. CDC. "Losing Weight — Healthy Weight." Recommends 1–2 lb/week. https://www.cdc.gov/healthyweight/losing_weight/
2. Hall KD, et al. (2011). "Quantification of the effect of energy imbalance on bodyweight." *Lancet* 378:826-837.
3. Kreitzman SN, et al. (1992). "Glycogen storage: illusions of easy weight loss." *Am J Clin Nutr* 56:292S-293S.
4. Helms ER, et al. (2014). "Evidence-based recommendations for natural bodybuilding contest preparation." *J Int Soc Sports Nutr* 11:20.

---

### How to Lose Weight Without Exercise (The Honest Guide, 2026)

URL: https://www.mycallie.app/en/blog/how-to-lose-weight-without-exercise
Markdown: https://www.mycallie.app/md/en/blog/how-to-lose-weight-without-exercise.md
Published: 2026-06-30 · Updated: 2026-06-30
Tags: how to lose weight without exercise, lose weight without gym, weight loss diet, calorie deficit

If the gym isn't happening — bad knees, no time, no interest — you can still lose weight. In fact, diet is the bigger lever anyway. Here's the honest, no-fluff playbook.

## The 30-second answer

**Diet drives about 80% of weight loss. You absolutely can lose weight without exercise — all you need is a sustained calorie deficit.** The reason it works: it's far easier to *not eat* 500 calories than to *burn* 500 calories. A candy bar takes 30 seconds to eat and an hour of walking to burn off. Everything below is a tool to reach that deficit without feeling miserable.

[Interactive calorie deficit calculator — use it on the page.]

## Why diet beats exercise for weight loss

This isn't anti-exercise (movement is great for your health, heart, and mood). It's about math:

- A 75 kg person burns roughly **300 kcal** in a 45-minute gym session.
- A single muffin, latte, and handful of nuts can easily *add* 600 kcal.
- You can undo an hour of exercise in two minutes of eating.

That asymmetry is why **"you can't out-train a bad diet"** is true — and why you can lose weight with no training at all, as long as intake stays below burn. Find your burn first with the calculator above, then aim **300–500 kcal below it.**

## The 6 levers that work without a gym

### 1. Cut liquid calories first

This is usually the single highest-impact change. Sodas, juices, sweetened coffees, smoothies, and alcohol deliver hundreds of calories without filling you up — your body barely registers liquid calories as food.

- A large sweetened coffee: 300–500 kcal
- A glass of juice: 120 kcal
- Two beers: 300 kcal

Swap to water, sparkling water, black coffee, or tea and many people remove 300–600 kcal/day without touching their meals.

### 2. Make protein the center of every plate

Protein is the most filling macronutrient and protects muscle while you lose. Aim for 1.6–2.2 g/kg bodyweight. Build meals around eggs, chicken, fish, Greek yogurt, tofu, legumes, or lean meat, and hunger drops on its own. ([Protein calculator.](/blog/protein-calculator))

### 3. Fill half your plate with fiber

Vegetables, fruit, beans, and whole grains are high-volume and low-calorie. Fiber slows digestion and triggers fullness, so you eat fewer calories without counting every bite. The "half plate veg, quarter protein, quarter carbs" rule does a lot of the deficit work automatically.

### 4. Use the 80% rule

Stop eating when you're about 80% full, not stuffed. Fullness signals lag ~20 minutes behind your stomach. Eat slowly, put the fork down between bites, and you'll naturally take in less.

### 5. Move more without "exercising" (NEAT)

Non-exercise activity — walking, standing, taking stairs, cleaning, fidgeting — can burn **hundreds of calories a day**, often more over a week than a few gym sessions. None of it is "working out":

- A 20-minute walk after each meal
- Standing or pacing on calls
- Parking farther, taking stairs
- A daily step target (even 7,000–8,000 steps moves the needle)

### 6. Protect sleep and manage stress

Short sleep raises ghrelin (hunger) and lowers leptin (fullness), and high cortisol drives cravings and water retention. People who sleep 5 hours eat measurably more than those who sleep 8. Fixing sleep can be as effective as a diet rule.

## A realistic, no-exercise day

| Lever | Action | Approx. deficit |
|---|---|---|
| Liquid calories | Black coffee instead of two sweet lattes | −400 kcal |
| Protein + fiber | Build meals to stay full on fewer calories | −200 kcal eaten |
| 80% rule | Stop before stuffed at dinner | −150 kcal |
| NEAT | Three 15-min walks + stairs | +250 kcal burned |

Stacked, these create a comfortable 500–800 kcal/day deficit — without a single squat.

## What to expect

A sustainable rate is **0.5–1 lb (0.25–0.5 kg) per week.** Faster usually means losing muscle and water, and it rebounds. Without exercise, **tracking accuracy matters even more** — you don't have a workout buffer for sloppy estimates, so honest logging is the difference between progress and a stall.

Without exercise it's tempting to cut very low. Going below your BMR backfires — energy crashes, NEAT drops (you move less without realizing), and you rebound. A moderate deficit you can hold for months beats an aggressive one you quit in two weeks.

## How Callie makes diet-only weight loss work

Since diet is doing 100% of the work, accuracy is everything — and that's exactly the friction Callie removes. Snap a photo of your plate, say what you ate, or type it, and Callie logs it in seconds with no database hunting. It sets a deficit matched to your timeline, catches the liquid calories and hidden oils that stall diet-only progress, and gives you BMI-matched daily plans so you know what to eat, not just what to avoid.

## Related reading

- [How Many Calories Do I Burn a Day? (TDEE)](/blog/maintenance-calories-tdee-calculator) — set your starting point.
- [How Many Calories Should I Eat to Lose Weight?](/blog/how-many-calories-to-lose-weight) — dial in your deficit.
- [How to Track Calories Accurately](/blog/how-to-track-calories-accurately) — the accuracy that diet-only loss depends on.
- [Why Have I Stopped Losing Weight?](/blog/weight-loss-plateau) — if the scale stalls.

## Sources

1. Mayo Clinic. "Weight loss: 6 strategies for success." https://www.mayoclinic.org/healthy-lifestyle/weight-loss/in-depth/weight-loss/art-20047752
2. MD Anderson Cancer Center. "How to lose weight without exercise." https://www.mdanderson.org/cancerwise/how-to-lose-weight-without-exercise.h00-159694389.html
3. Spiegel K, et al. (2004). "Sleep curtailment results in increased hunger and appetite." *Ann Intern Med* 141:846-850.
4. Levine JA (2002). "Non-exercise activity thermogenesis (NEAT)." *Best Pract Res Clin Endocrinol Metab* 16(4):679-702.

---

### How to Track Calories Accurately (And Stop Under-Logging by 40%)

URL: https://www.mycallie.app/en/blog/how-to-track-calories-accurately
Markdown: https://www.mycallie.app/md/en/blog/how-to-track-calories-accurately.md
Published: 2026-06-30 · Updated: 2026-06-30
Tags: how to track calories accurately, calorie tracking mistakes, logging food accurately, portion sizes

If your calorie deficit "isn't working," the problem usually isn't your metabolism or the food database — it's that you're eating more than you're logging, without realizing it. Here's how to fix the accuracy of your tracking.

## The 30-second answer

**The average person under-logs by 20–40%.** That's not lying — it's forgotten oils, eyeballed portions, untracked drinks, and bites that never get entered. To track accurately: **weigh calorie-dense foods in grams, log before or as you eat (not from memory at night), and count everything with calories — including liquids and "tastes."** Get those right and your numbers finally match reality.

## The under-logging problem is bigger than you think

A landmark *New England Journal of Medicine* study found that people who believed they ate ~1,000 kcal/day were actually eating closer to 2,000 — under-reporting by **nearly 50%** — and they weren't lying; they genuinely didn't perceive it. Decades of research since confirm a typical **20–40% gap** between reported and actual intake.

This single fact explains most "I'm in a deficit but not losing weight" cases. The deficit on paper is real; the deficit in your kitchen isn't.

## The 4 hidden-calorie categories

Almost all of the gap comes from four places:

| Category | Typical hidden cost | Why it's missed |
|---|---|---|
| **Cooking oils & fats** | 100–300 kcal/meal | A "tablespoon" of oil is usually two; butter on the pan goes uncounted |
| **Liquid calories** | 150–600 kcal/day | Juice, soda, alcohol, sweet coffee don't feel like "food" |
| **Condiments & sauces** | 50–200 kcal/meal | Dressing, mayo, ketchup, peanut sauce add up invisibly |
| **Bites, licks & tastes** | 100–300 kcal/day | Tasting while cooking, kids' leftovers, "just one chip" |

If you only fix one thing, **measure your cooking oil.** It's the most calorie-dense thing in most kitchens and the most consistently undercounted.

## Weigh in grams, not cups or handfuls

Visual portion estimates are off by **20–50%**, and the error almost always skews toward *under*-counting. The fix is a cheap digital food scale:

- Weigh in **grams**, not volume. "One cup of rice" varies wildly; 150 g doesn't.
- Weigh the **calorie-dense stuff** especially: oils, nut butters, cheese, grains, pasta, meat. A 30% error on broccoli is ~5 kcal; a 30% error on peanut butter is ~100 kcal.
- Weigh **raw where possible**, and pick the database entry that matches (cooked vs raw differ a lot for rice, pasta, meat).

You don't have to weigh forever. A few weeks recalibrates your eye, so your estimates get genuinely good.

## Log before you eat, not from memory

When you log matters as much as how:

1. **Pre-log the day.** Sketch tomorrow's meals tonight. This cuts tracking error by roughly half — you can't forget an entry, and you decide portions before hunger hijacks the decision.
2. **Log as you eat** if you can't pre-log. Real-time beats reconstruction.
3. **Never rely on end-of-day memory.** This is where the bites, the second helping, and the office snacks vanish.

Accurate tracking isn't about a perfect day — it's about an honest one. Logging the cookie you didn't plan is more valuable than pretending it didn't happen. The goal is a true picture you can act on, not a clean-looking diary.

## Other accuracy traps

- **Eating exercise calories back in full.** Wearables overestimate burn 20–40%. Eat back at most ~50%, less for walking.
- **Trusting restaurant menu numbers.** Real plates exceed listed calories by ~20% on average. Round up.
- **Database entries with no weight.** "1 chicken breast" is meaningless — sizes vary 2×. Use gram entries.
- **Cooked vs raw mismatches.** 100 g of raw chicken ≠ 100 g cooked. Pick the right entry.

## Why photo and voice logging are more accurate in practice

Here's the counterintuitive part: the most "precise" method on paper — manual database search — is often the *least* accurate in practice, because its friction makes you skip, guess, and give up. Every "I'll log it later" is a future omission.

Photo and voice logging flip this. Snapping a plate or saying "two eggs, toast with butter, and a flat white" takes seconds, so you actually do it **every time** — and capturing the meal in the moment means you don't forget the butter or the coffee. Modern AI lands within ~10–25% on typical meals, comparable to careful manual entry. The accuracy win is mostly about **adherence**: the best tracking method is the one you'll keep using.

## How Callie keeps you accurate

Callie is built around the under-logging problem. Photograph your plate and it estimates portions and calories; or just say what you ate and it logs it. There's a **review-before-confirm step** so you can correct the estimate in one tap — accuracy without the data-entry grind. Because logging takes seconds, you capture the oils, drinks, and bites that wreck a deficit on every other app. And it works in your language, so describing food never trips you up.

## Related reading

- [Calorie Deficit but Not Losing Weight? 11 Real Reasons](/blog/calorie-deficit-not-losing-weight) — under-logging is reason #1.
- [How to Track Calories From a Photo](/blog/how-to-track-calories-from-a-photo) — how AI photo logging works.
- [How Many Calories Should I Eat to Lose Weight?](/blog/how-many-calories-to-lose-weight) — set the target you're tracking against.
- [How Many Calories Do I Burn a Day? (TDEE)](/blog/maintenance-calories-tdee-calculator)
- [AI Calorie Apps Underestimate Meals by up to 345 Calories](/blog/ai-calorie-app-accuracy-study-2026) — what the 2026 NIH study found, and why review-before-confirm matters.

## Sources

1. Lichtman SW, et al. (1992). "Discrepancy between self-reported and actual caloric intake and exercise in obese subjects." *N Engl J Med* 327(27):1893-1898.
2. Champagne CM, et al. (2002). "Energy intake underreporting." *J Am Diet Assoc* 102(10):1428-1432.
3. Shcherbina A, et al. (2017). "Accuracy in Wrist-Worn, Sensor-Based Measurements of Energy Expenditure." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/
4. Urban LE, et al. (2011). "Accuracy of stated energy contents of restaurant foods." *JAMA* 306(3):287-293.

---

### Intermittent Fasting vs Calorie Counting: Which Actually Works? (2026)

URL: https://www.mycallie.app/en/blog/intermittent-fasting-vs-calorie-counting
Markdown: https://www.mycallie.app/md/en/blog/intermittent-fasting-vs-calorie-counting.md
Published: 2026-06-30 · Updated: 2026-06-30
Tags: intermittent fasting vs calorie counting, is intermittent fasting better, 16:8 weight loss, calorie counting

It's the most-searched diet debate of the decade, and most articles get it wrong by picking a side. Here's the honest, research-backed answer.

## The 30-second answer

**They work about equally well — because they work the same way.** Both intermittent fasting (IF) and calorie counting cause weight loss by creating a **calorie deficit**. When researchers match total calories and protein, the difference in fat loss between the two is small and inconsistent. The real question isn't which is superior; it's **which one you'll actually stick to.**

## What the research actually shows

Multiple randomized trials and reviews have put the two head-to-head:

- A year-long trial in the *New England Journal of Medicine* found **no significant difference** in weight loss between time-restricted eating and daily calorie restriction.
- A 2024 review in the *Annals of Internal Medicine* concluded that for most people the two are comparable; adherence and food quality mattered more than the method itself.
- Harvard Health's summary is blunt: fasting "works" because it's a tool to reduce calorie intake — not because of a unique metabolic effect.

The takeaway: **fasting is a strategy for eating less, not a loophole around the calorie equation.**

## Head-to-head comparison

| Factor | Intermittent fasting | Calorie counting |
|---|---|---|
| **How it cuts calories** | Shrinks the *window* you eat in | Caps the *amount* you eat |
| **Precision** | Low — you don't know your intake | High — you know your number |
| **Decision load** | Low (one rule: when to eat) | Moderate (log meals) |
| **Flexibility** | Rigid timing, flexible food | Flexible timing, tracked food |
| **Protein target** | Harder (short window) | Easier (planned across day) |
| **Best for** | People who hate tracking | People who want control & data |
| **Main failure mode** | Overeating in the window | Tracking fatigue / fixation |

## Where intermittent fasting wins

- **Fewer decisions.** "Don't eat before noon" is one rule. For people who find tracking exhausting, that simplicity drives real adherence.
- **No logging.** You can lose weight without ever touching an app, *if* shrinking your window naturally reduces intake.
- **Routine appetite control.** Many people find a defined window curbs evening grazing.

## Where calorie counting wins

- **It works for everyone.** Fasting only helps if a shorter window makes you eat less. Counting works regardless of *when* you eat.
- **Precision and flexibility.** Want pizza tonight? Counting tells you exactly how to fit it. Fasting just tells you to eat it before 8 p.m.
- **Protein and muscle.** Spreading protein across the day is easier when you're planning intake, which protects muscle in a deficit.
- **It catches the hidden calories.** The cooking oil, the latte, the "bites" — the things that stall a fast-only approach because nothing is being measured.

## The case for doing both

You don't have to choose. The highest-adherence approach for many people is **fasting for structure + loose tracking for accuracy**:

1. Pick an eating window you can keep (16:8 is the common starting point — eat within 8 hours).
2. Loosely log what you eat in that window — enough to stay honest, not enough to obsess.
3. Hit a protein target in every meal so the short window doesn't cost you muscle.

This gives you fasting's "one simple rule" *and* counting's safety net against the invisible calories that quietly undo progress. You don't need to track forever — even a few weeks recalibrates your eye for portions.

Don't ask "which is scientifically better?" Ask "which will I still be doing in six months?" Adherence beats optimization every time. If you've tried and abandoned one, switch — that's data, not failure.

## Common mistakes with each

**Fasting:**
- Treating the window as a free-for-all and eating past maintenance.
- Crushing training performance with too tight a window around workouts.
- Under-eating protein because there's "no time."

**Counting:**
- Under-logging by 20–40% (the universal tracking error) and wondering why the scale won't move.
- Letting precision tip into anxiety. Tracking is a tool, not a moral scorecard.

## How Callie fits either approach

Callie doesn't force a method on you. If you fast, log only your eating-window meals — by photo, voice, or text in a couple of seconds — and Callie keeps protein and calories honest without making tracking a chore. If you count, Callie sets your target and adjusts it as you go. Either way, the friction that makes people quit tracking is gone, which is the whole point: **the method only works if you keep doing it.**

## Related reading

- [Intermittent Fasting: A Practical Guide](/guides/intermittent-fasting) — windows, protocols, and who should skip it.
- [How Many Calories Should I Eat to Lose Weight?](/blog/how-many-calories-to-lose-weight) — set your deficit target.
- [How to Track Calories Accurately](/blog/how-to-track-calories-accurately) — fix the under-logging that stalls a fast-only approach.
- [Calorie Deficit but Not Losing Weight? 11 Real Reasons](/blog/calorie-deficit-not-losing-weight)

## Sources

1. Lowe DA, et al. (2020). "Effects of Time-Restricted Eating on Weight Loss (TREAT trial)." *JAMA Intern Med* 180(11):1491-1499.
2. Liu D, et al. (2022). "Calorie Restriction with or without Time-Restricted Eating in Weight Loss." *N Engl J Med* 386:1495-1504.
3. Harvard Health Publishing. "Fasting vs. calorie restriction for weight loss: Which one wins?" https://www.health.harvard.edu/staying-healthy/fasting-vs-calorie-restriction-for-weight-loss-which-one-wins
4. Annals of Internal Medicine (2024). Review of intermittent fasting vs continuous calorie restriction.

---

### How Many Calories Do I Burn a Day? (TDEE & Maintenance Calculator)

URL: https://www.mycallie.app/en/blog/maintenance-calories-tdee-calculator
Markdown: https://www.mycallie.app/md/en/blog/maintenance-calories-tdee-calculator.md
Published: 2026-06-30 · Updated: 2026-06-30
Tags: maintenance calories, TDEE calculator, how many calories do I burn a day, BMR vs TDEE

If you want one number to build everything else on — losing, gaining, or holding — it's your **maintenance calories**, also called TDEE. Here's how to find it, why it differs from the formula, and how to use it.

## The 30-second answer

**Your maintenance calories (TDEE) are the total you burn in a day.** Eat that number and your weight stays flat. For most adults it lands between 1,800 and 2,800 kcal. Use the calculator below for your personal estimate, then read on for why the formula is only a starting point.

[Interactive calorie deficit calculator — use it on the page.]

## BMR vs TDEE: what the calculator is actually doing

Two numbers do all the work:

- **BMR (basal metabolic rate)** — what your body burns at complete rest, just to keep your heart, brain, liver, and kidneys running. For most people this is **60–70% of total daily burn**.
- **TDEE (total daily energy expenditure)** — BMR plus everything else you do. That's the number you eat against.

TDEE breaks down into four parts:

| Component | What it is | Share of TDEE |
|---|---|---|
| BMR | Resting/organ function | 60–70% |
| NEAT | Non-exercise movement: walking, standing, fidgeting, chores | 15–30% |
| TEF | Thermic effect of food (energy to digest) | ~10% |
| EAT | Exercise (deliberate training) | 5–15% for most people |

The surprise for most people: **NEAT, not the gym, explains most of the difference** between a "fast metabolism" and a "slow" one. Two people of identical size can differ by 400+ kcal/day purely in how much they move around outside of workouts.

## Maintenance calories — quick reference

These use the Mifflin-St Jeor equation at a **moderate activity multiplier (1.55)**. Find the row closest to you, then adjust with the activity guide below.

### Women

| Age | Weight (kg) | Height (cm) | Maintenance (kcal) |
|---|---|---|---|
| 25 | 55 | 160 | 1,890 |
| 25 | 70 | 168 | 2,120 |
| 35 | 60 | 162 | 1,930 |
| 35 | 75 | 170 | 2,150 |
| 45 | 65 | 163 | 1,940 |
| 45 | 80 | 170 | 2,160 |
| 55 | 70 | 162 | 1,950 |

### Men

| Age | Weight (kg) | Height (cm) | Maintenance (kcal) |
|---|---|---|---|
| 25 | 70 | 175 | 2,680 |
| 25 | 85 | 182 | 2,930 |
| 35 | 75 | 176 | 2,660 |
| 35 | 90 | 183 | 2,900 |
| 45 | 80 | 177 | 2,650 |
| 45 | 95 | 184 | 2,880 |
| 55 | 85 | 176 | 2,620 |

### Adjust for your activity level

| If you are... | Multiply BMR by | Or adjust the table by |
|---|---|---|
| Sedentary (desk job, little exercise) | 1.2 | −20% |
| Lightly active (1–3 days/week) | 1.375 | −10% |
| Moderately active (3–5 days/week) | 1.55 | baseline |
| Very active (6–7 days/week) | 1.725 | +10% |
| Athlete / physical job | 1.9 | +20% |

Most people overestimate this. Studies repeatedly show people rate themselves "moderately active" when they're really "lightly active." When in doubt, pick the lower bracket — you can always add calories if the scale isn't moving.

## Why your real burn differs from the number

The Mifflin-St Jeor equation is the most accurate general formula available, but it was built on population averages. Your number can drift from it for real reasons:

- **Muscle mass** — more lean tissue raises BMR. Two people at 75 kg can differ by 150+ kcal based on body composition alone. (If you know your body-fat %, the Katch-McArdle formula is more precise.)
- **NEAT variability** — fidgeters and pacers burn hundreds more calories a day than still people, often without noticing.
- **Adaptive thermogenesis** — after sustained dieting, BMR drops a little *below* what your new body size predicts. This is real but small (typically 50–100 kcal), and reverses when you eat at maintenance again.
- **Age and hormones** — BMR declines slowly with age, largely tracking muscle loss.

This is why the formula is a **starting point, not a verdict**. The scale is the real calorimeter.

## How to find your true maintenance in 3 weeks

1. **Eat at your calculated TDEE** for 2–3 weeks, logging consistently.
2. **Weigh daily, take the weekly average.** Daily weight swings ±1 kg from water, sodium, and glycogen — the average cancels the noise.
3. **Read the trend:**

| Weekly average | What it means | Action |
|---|---|---|
| Holding steady | Your estimate is correct | This is your true maintenance |
| Slowly rising | Real TDEE is lower than estimated | Subtract ~150 kcal |
| Slowly falling | Real TDEE is higher (or you under-logged) | Add ~150 kcal |

After 3 weeks you'll have a maintenance number grounded in *your* data — far more reliable than any equation.

## A word on "starvation mode"

There is no switch that makes your body stop losing weight if you eat too little. What's real is **metabolic adaptation**: as you get smaller and diet longer, your burn drifts down, which shrinks your deficit over time. The fix isn't to eat even less — it's to know your *current* maintenance and keep a sensible gap below it. If weight loss stalls completely, see [why you've stopped losing weight](/blog/weight-loss-plateau).

## Don't trust your smartwatch's calorie number

Wrist-worn trackers are good at counting steps and heart rate, but bad at calories. They overestimate exercise burn by **20–40%** on average ([Shcherbina et al., 2017](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/)). If you eat back the "active calories" your watch reports, you can quietly erase the deficit you're trying to create. Use the watch for activity trends, not as your eating target.

## How Callie handles this for you

Tell Callie your age, height, weight, and activity, and it sets your maintenance and goal targets using Mifflin-St Jeor — then **re-tunes them automatically every two weeks** based on your real weight trend, so you never have to recalculate after losing a few kilos. Because you log meals by photo, voice, or text instead of searching a database, your intake data is accurate enough for that auto-tuning to actually work.

## Related reading

- [Calorie Deficit Calculator & Complete Guide](/guides/calorie-calculator) — turn your maintenance number into a fat-loss target.
- [How Many Calories Should I Eat to Lose Weight?](/blog/how-many-calories-to-lose-weight) — the deficit side of the equation.
- [How Long Does It Take to Lose Weight?](/blog/how-long-does-it-take-to-lose-weight) — realistic timelines built on your TDEE.
- [Why Have I Stopped Losing Weight?](/blog/weight-loss-plateau) — what to do when maintenance drifts down.

## Sources

1. Mifflin MD, et al. (1990). "A new predictive equation for resting energy expenditure in healthy individuals." *Am J Clin Nutr* 51(2):241-247.
2. Shcherbina A, et al. (2017). "Accuracy in Wrist-Worn, Sensor-Based Measurements of Heart Rate and Energy Expenditure in a Diverse Cohort." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/
3. Levine JA (2002). "Non-exercise activity thermogenesis (NEAT)." *Best Pract Res Clin Endocrinol Metab* 16(4):679-702.
4. Rosenbaum M, Leibel RL (2010). "Adaptive thermogenesis in humans." *Int J Obes* 34:S47-S55.

---

### Why Have I Stopped Losing Weight? Breaking a Weight-Loss Plateau (2026)

URL: https://www.mycallie.app/en/blog/weight-loss-plateau
Markdown: https://www.mycallie.app/md/en/blog/weight-loss-plateau.md
Published: 2026-06-30 · Updated: 2026-06-30
Tags: weight loss plateau, stopped losing weight, why am I not losing weight, diet break

You were losing steadily, then the scale froze. Before you slash calories or blame your metabolism, work through this diagnostic — most "plateaus" have a fixable, unglamorous cause.

## The 30-second answer

**A real plateau means your weekly average weight hasn't moved in 3+ weeks despite a genuine deficit.** It happens because (1) a smaller body burns fewer calories, so your old deficit shrank to nothing, and (2) tracking quietly got looser. The fix is rarely "eat less first" — it's **recalculate maintenance for your new weight, tighten your logging, and check protein, sleep, and steps.** About 85% of dieters hit this; it's a stage, not a failure.

## First: is it actually a plateau?

Most stalls under three weeks aren't real. Daily weight swings ±1–2 kg from water, sodium, glycogen, digestion, and (for women) cycle phase. A single hard workout can hold water for days.

**Diagnose it properly:**

| Signal | Likely a real plateau? |
|---|---|
| Stall under 2 weeks | No — almost certainly water/noise |
| 2–3 weeks, weekly average flat | Maybe — keep watching |
| 3+ weeks, weekly average flat, tracking tight | Yes — time to act |
| Waist/clothes still changing | No — you're losing fat, holding water |

Judge by your **weekly average**, not any single morning. If you only weigh once a week, you can't tell signal from noise.

## Why real plateaus happen

When the scale genuinely stops, it's some mix of these:

- **Your maintenance dropped.** A 90 kg body burns more than a 78 kg body. After losing 12 kg, the deficit that once melted weight off may now equal your new maintenance. **This is the #1 cause.**
- **Tracking drifted.** Portions creep up, "bites" stop getting logged, the oil goes uncounted. People under-log by 20–40% on average — and it gets worse the longer you've been at it.
- **Exercise burn overestimated.** Wearables inflate calorie burn 20–40%. Eating those back erases the deficit.
- **Metabolic adaptation.** After sustained dieting, BMR drops slightly below predicted, and NEAT falls (you move less, often unconsciously) — together shrinking the gap.
- **Muscle loss.** Dieting without enough protein or resistance training costs muscle, which lowers your burn further.

## The fix, in order

Work top to bottom. Don't jump to "eat less" — it's near the bottom for a reason.

1. **Recalculate maintenance for your current weight.** Use your new numbers, not the ones from 10 kg ago. ([Find your TDEE here.](/blog/maintenance-calories-tdee-calculator)) Reset your deficit from the new maintenance.
2. **Tighten tracking for 10 days.** Weigh and log everything — oils, drinks, sauces, bites. Pre-log the day if you can. This alone restarts more plateaus than any other step.
3. **Check protein.** Aim for 1.6–2.2 g/kg bodyweight. Protein protects muscle (keeping your burn up) and controls hunger. ([Protein calculator.](/blog/protein-calculator))
4. **Raise NEAT.** Add 2,000–3,000 daily steps. NEAT is the first thing to fall during a diet and the easiest burn to reclaim.
5. **Audit sleep and stress.** Poor sleep and high cortisol raise hunger hormones and water retention, masking fat loss and sabotaging adherence.
6. **Only then, adjust intake** — and prefer a small cut (100–150 kcal) or more activity over a drastic slash.

## The counterintuitive fix: take a diet break

If you've been dieting hard for months, the best move might be to **stop dieting for one to two weeks** — eat at maintenance, not below.

The MATADOR study found that dieters who alternated two weeks of deficit with two weeks at maintenance lost *more* fat and regained less than those who dieted continuously. A planned break:

- Restores leptin and other diet-regulating hormones.
- Brings NEAT back up (you move normally again).
- Resets adherence and willpower so you don't quit entirely.

The key word is **planned**. A diet break is a strategy with a start and end date — not "I'll take a few days off" that becomes three months.

The most damaging response to a stall isn't eating at maintenance for two weeks — it's getting discouraged and abandoning everything. A consistent maintenance phase is part of the plan. A cycle of give-up-and-rebound is what actually undoes progress.

## How Callie helps you push through

- **Auto-recalculated targets.** As you lose weight, Callie updates your maintenance and deficit every two weeks — so the #1 plateau cause (eating at your *old* maintenance) never silently happens.
- **Accurate, low-friction logging.** Photo and voice logging make it hard to under-count, which catches the tracking drift that causes most stalls.
- **Streak-preserving cheat days and diet breaks.** Callie lets you take a planned maintenance day or break *without breaking your streak* — so a strategic pause feels like progress, not failure, and you actually stick with it.

## Related reading

- [How Many Calories Do I Burn a Day? (TDEE)](/blog/maintenance-calories-tdee-calculator) — recalculate for your new weight.
- [Calorie Deficit but Not Losing Weight? 11 Real Reasons](/blog/calorie-deficit-not-losing-weight) — the deep-dive diagnostic.
- [How to Track Calories Accurately](/blog/how-to-track-calories-accurately) — fix the under-logging behind most stalls.
- [Protein Calculator](/blog/protein-calculator) — protect the muscle that keeps your metabolism up.

## Sources

1. Sainsbury A, et al. (2018). "Effect of intermittent dieting on fat loss (MATADOR study)." *Int J Obes* 42:129-138.
2. Mayo Clinic. "Getting past a weight-loss plateau." https://www.mayoclinic.org/healthy-lifestyle/weight-loss/in-depth/weight-loss-plateau/art-20044615
3. Rosenbaum M, Leibel RL (2010). "Adaptive thermogenesis in humans." *Int J Obes* 34:S47-S55.
4. "Management of Weight Loss Plateau." StatPearls, NCBI Bookshelf. https://www.ncbi.nlm.nih.gov/books/NBK576400/

---

### Calorie Deficit but Not Losing Weight? 11 Real Reasons (2026)

URL: https://www.mycallie.app/en/blog/calorie-deficit-not-losing-weight
Markdown: https://www.mycallie.app/md/en/blog/calorie-deficit-not-losing-weight.md
Published: 2026-05-27 · Updated: 2026-05-27
Tags: calorie deficit not losing weight, weight loss plateau, stuck at same weight, deficit not working

If your weight is stuck and you think you're in a deficit, one of these is the answer. The diagnostic flowchart below points you at the most likely cause in three questions; the eleven sections after that explain each in detail with the specific fix.

## Diagnostic flowchart

**Question 1:** Has your 7-day average weight been flat for 3+ weeks (not 3+ days)?

- **No** → You're probably not actually stuck. Body weight has 1–2% daily variance; a flat week is noise. Wait two more weeks before changing anything. Skip to [reason #7](#7-youre-comparing-single-weigh-ins-not-7-day-averages).
- **Yes** → Continue.

**Question 2:** Are you logging every food on a kitchen scale before eating it?

- **No** → Start with [reason #1](#1-youre-under-logging-the-biggest-one). This is the cause about 70% of the time.
- **Yes (kitchen scale, before eating)** → Continue.

**Question 3:** Are you eating back exercise calories from a fitness watch?

- **Yes** → Start with [reason #2](#2-your-watch-is-overestimating-exercise-burn).
- **No** → Work through [reasons #3 through #11](#3-your-real-tdee-is-lower-than-the-formula-says).

The rest of this article walks through each reason with the specific fix.

## 1. You're under-logging (the biggest one)

This is the most common cause by a wide margin. The classic Lichtman et al. study published in *The New England Journal of Medicine* (1992) compared self-reported intake against doubly-labeled water measurements in subjects who described themselves as diet-resistant. The result: they under-reported their intake by roughly 47% on average. Follow-up research across general populations consistently shows 20–40% under-reporting.

**Why it happens, even when you're trying:**

- Eyeballed portions are systematically biased toward smaller estimates.
- Cooking oils and dressings get forgotten — a tablespoon of olive oil is 120 kcal.
- "Tastes" while cooking — a few bites of pasta as you stir adds 50–100 kcal nobody logs.
- Drinks (juice, latte, beer, wine) are skipped because they don't "feel" like food.
- Restaurant portions are 1.5–2× the "standard" serving the app uses.

**The fix:**

- Use a kitchen scale for two weeks. Weigh every food before you eat it.
- Log it *before* eating, not after — log-after-eating loses about 25% of items.
- Count drinks. Every drink. Lattes alone can add 600 kcal/week.
- Use an AI photo logger like [Callie's photo logging](/blog/how-to-track-calories-from-a-photo) for meals you can't weigh — the [review-and-edit step](/guides/ai-calorie-tracking) catches portion misjudgments before they bake into your average.

## 2. Your watch is overestimating exercise burn

Consumer wrist-worn fitness watches overestimate exercise calorie expenditure by 20–40% for most activities, with the worst errors in strength training and HIIT. A watch that says "you burned 600 kcal" on a hard gym session is closer to 350–450 in reality.

The trap: you eat back the inflated number. Now your deficit is gone — or worse, you're in a surplus on training days.

**The fix:**

- For walking and easy cardio, eat back only ~25% of what your watch reports.
- For moderate cardio (running, cycling), eat back ~50%.
- For strength training, don't eat back watch calories at all — assume ~5 kcal/min for moderate sessions and stop there.
- Better: pick a daily target that already assumes some exercise, and don't add more on top of it.

## 3. Your real TDEE is lower than the formula says

Mifflin-St Jeor and other BMR formulas are validated to roughly ±10% for healthy adults. That means a formula spitting out "1,700 kcal BMR" could be 1,530 to 1,870 in reality. The error is bigger for:

- Very lean people (formulas under-estimate)
- People with high body-fat percentage (formulas over-estimate)
- Post-menopausal women (under-counted by most formulas)
- Anyone over 60 (TDEE drops faster than the formula tracks)

**The fix:**

- Track 14 days of carefully measured intake and 14 days of weight averages.
- Use the actual data: if you ate 1,600 kcal/day on average and the trend held flat, your TDEE is ~1,600 — not what the formula said.
- Drop another 200 kcal/day from the empirical number, not the formula number.

## 4. Water retention from new training

Starting a new strength or high-intensity training program causes muscles to store extra glycogen and water for 2–4 weeks. This can add 1–2 kg that masks real fat loss.

**The signal:** the scale is flat or up, but the mirror, waistband, and tape measurements are going in the right direction.

**The fix:** wait 3–4 weeks. The water settles, and the real fat loss becomes visible. Don't change calories in the meantime.

## 5. Glycogen and carb rebound

Each gram of stored glycogen binds about 3 grams of water. A high-carb day (cheat meal, refeed, restaurant pasta) replenishes glycogen and can add 1–2 kg overnight. None of it is fat — it's gone within 3–4 days on a normal pattern.

**The fix:** if your scale jumped after a high-carb day, ignore it. Look at the average over the next 7 days. If you bounce around weekly cheat days, weigh on the same day of the week (e.g. Friday before breakfast) for a comparable baseline.

## 6. You're not sleeping enough

Sleep restriction (under 6.5 hours/night) does two measurable things:

- Raises ghrelin (hunger hormone) and lowers leptin (satiety hormone). Studies put the appetite increase at roughly 200–400 kcal/day.
- Lowers NEAT (non-exercise activity thermogenesis) — you fidget less, take fewer steps, take the elevator.

Combined effect: a 400–500 kcal/day swing toward maintenance, often without you realizing it.

**The fix:** prioritize 7+ hours. If sleep is non-negotiable for life reasons, accept a slower deficit rather than pushing harder against a sleep-debt headwind.

## 7. You're comparing single weigh-ins, not 7-day averages

Body weight has 1–2% daily variance from:

- Sodium intake (each gram of sodium binds ~10 ml of water)
- Glycogen swings
- Bowel movements
- Menstrual cycle (women)
- Time of day, hydration status

A daily reading is essentially noise. The right measurement is the **7-day rolling average** — sum your last 7 daily weights, divide by 7.

**The fix:** weigh every morning, same time, same conditions. Track the 7-day average. Compare this week's average to four weeks ago — if it's down 0.4–0.8 kg, you're losing on schedule even if individual days look chaotic.

## 8. Hidden calories you've stopped counting

The biggest categories of forgotten kcal:

- **Cooking oils.** A tablespoon of any oil is 120 kcal. A "drizzle" is often 2 tablespoons.
- **Dressings and sauces.** A typical salad dressing serving is 80–150 kcal. Restaurant servings are usually 2–3× that.
- **Coffee additions.** Cream, sugar, syrups. A daily latte alone is 150–250 kcal × 365 days = 55,000–90,000 kcal/year.
- **Alcohol.** A glass of wine is 120–180 kcal. A pint of beer is 180–250 kcal.
- **"Tastes."** Bites of food while cooking, finishing kids' plates, samples at the grocery store.
- **Condiments.** Ketchup, BBQ sauce, peanut butter, hummus — 50–200 kcal per serving and most people don't log them.

**The fix:** for two weeks, log absolutely everything. Even a single grape. The point isn't to keep doing it forever — it's to find the hole in your normal tracking pattern.

## 9. Restaurant and packaged-food labels are off

US FDA regulations allow packaged food labels to be off by up to 20%. Restaurant calorie estimates are notoriously worse — chain-restaurant menus have been measured at ±30% from listed values in studies.

**The fix:** when eating out, default to ordering simple grilled-protein-plus-vegetable plates where the calorie estimate is harder to be wrong about. Avoid the order categories with the widest variance: pasta, salads with heavy dressing, casseroles, layered dishes, anything described as "creamy" or "loaded."

## 10. Menstrual cycle water weight

Hormonal cycles cause 0.5–2 kg of water-weight swing in many women, with the peak typically in the late luteal phase (5–7 days before period start). This can entirely mask 2–3 weeks of fat loss progress.

**The fix:** track weight in 28-day blocks, compared against the same week of the previous cycle. Week 3 of this cycle vs. Week 3 of last cycle is a real comparison; today vs. two weeks ago in different cycle phases is not.

## 11. Metabolic adaptation (long-diet plateau)

After 8–16 weeks of continuous deficit, TDEE drops by 5–15% more than the formulas predict. This is *adaptive thermogenesis* — the body downregulates several systems (thyroid output, NEAT, reproductive hormones) when it senses a sustained energy deficit. It's well-documented in long-term weight-loss research, including the famous *Biggest Loser* follow-up studies.

**The fix — a planned diet break:**

- Take 1–2 weeks at maintenance calories (your current TDEE — not your pre-diet TDEE).
- Eat normally; track only to make sure you're not blowing past maintenance.
- After the break, resume the deficit. Most people see the scale resume movement within 7–10 days of restarting.

This is not "giving up" or "cheating." It's the same tool elite physique competitors use — planned mini-rests during long cuts.

## Putting the diagnosis together

If you've worked through all 11 and the scale still won't move, the honest possibilities are:

1. **You're not actually in a deficit.** Two empirical weeks at strictly-weighed intake + flat 7-day weight average is the test. If 1,600 kcal × 14 days = flat weight, your TDEE is 1,600, not whatever the formula says.
2. **You need a diet break.** Eight to twelve weeks of continuous deficit is enough to require one.
3. **There's a non-calorie cause.** New medications (especially antidepressants, antipsychotics, beta-blockers, insulin, prednisone, hormonal contraceptives) can cause weight gain or water retention independent of intake. Thyroid problems, PCOS, and Cushing's syndrome can also shift the math. If you've ruled out the eleven reasons above and added new medication or a recent diagnosis, talk to a doctor.

## How tracking accuracy changes the picture

The single highest-leverage change for someone stuck in a deficit is usually **better food tracking**, not eating less. If your under-logging is even 20% — well within the documented range — then a "1,600 kcal/day" log is really 1,920, which is maintenance for most active adults.

This is the specific failure mode AI photo loggers are designed to fix:

- Photo logging captures the whole plate, including the things you'd skip when typing meal-by-meal.
- A [review-and-edit step before confirming](/blog/how-to-track-calories-from-a-photo) lets you correct portion size on dishes the model would otherwise under-estimate.
- Voice logging is faster than manual entry, which reduces abandonment — the diet works for the months it has to work, not just the first three weeks.

See [Callie's photo and voice logging](/guides/ai-calorie-tracking) for how this is implemented.

## When to stop adjusting and accept the data

Sometimes the answer is: the diet target is slower than you'd like, but it's working. 0.5% of bodyweight per week is sustainable; 1%/week is the upper end without losing muscle. A 70 kg adult losing 0.35 kg/week is doing exactly what the math allows for and shouldn't push harder.

If your 7-day average is dropping by 0.3–0.7 kg/week, **you are losing weight as fast as is safe.** Don't cut more. Don't add cardio in panic. Keep going.

The two highest-leverage diagnostics: (1) weigh food for 14 days, log it before eating; (2) track 7-day weight averages, not daily readings. Those two changes alone resolve roughly 80% of "I'm in a deficit but not losing" complaints.

## Related reading

- [How many calories should I eat to lose weight?](/blog/how-many-calories-to-lose-weight) — the daily targets and the math behind them.
- [Macro calculator with cheat day adjuster](/blog/macro-calculator-with-cheat-day) — what to do when a single day blows up.
- [Protein calculator](/blog/protein-calculator) — protect muscle in a deficit by hitting a real protein floor.
- [How to track calories from a photo](/blog/how-to-track-calories-from-a-photo) — the under-logging fix that's actually sustainable.

---

### GLP-1 Diet Plan: What to Eat on Ozempic, Wegovy & Mounjaro (2026)

URL: https://www.mycallie.app/en/blog/glp-1-diet-plan-ozempic-wegovy-mounjaro
Markdown: https://www.mycallie.app/md/en/blog/glp-1-diet-plan-ozempic-wegovy-mounjaro.md
Published: 2026-05-27 · Updated: 2026-05-27
Tags: GLP-1 diet plan, what to eat on Ozempic, Wegovy meal plan, Mounjaro food list, Zepbound diet

This article summarizes published research and commonly recommended nutrition principles for adults on GLP-1 therapy. It is not medical advice and does not replace guidance from your prescriber, endocrinologist, or registered dietitian. Always confirm changes to your eating pattern with the clinician managing your medication.

## What this article is

A research summary of what the published guidance and clinical experience suggest about eating well on **semaglutide** (Ozempic, Wegovy) and **tirzepatide** (Mounjaro, Zepbound). It covers nutrition principles only — not dosing, prescription decisions, or side-effect management beyond food choices.

## Why nutrition matters more on a GLP-1 than off it

GLP-1 receptor agonists work primarily through two effects:

1. **Central appetite suppression.** They act on hunger and satiety pathways in the brain, sharply reducing the drive to eat.
2. **Delayed gastric emptying.** Food sits in the stomach longer, which extends satiety but also raises the risk of nausea, reflux, and bloating after large meals.

Both effects compound to drop daily calorie intake by roughly 20–35% in clinical trials. That is great for weight loss — but it creates two distinct risks the published literature is consistent about:

- **Muscle loss.** Trials of both semaglutide and tirzepatide have reported that 25–40% of the total weight lost can be lean mass when no protein or training intervention is layered on. (See manufacturer labels and the STEP and SURMOUNT trial programs.)
- **Nutrient gaps.** When intake drops, micronutrient density — protein, calcium, iron, B-vitamins, fiber — has to rise per calorie to keep you healthy. Most people don't adjust this automatically.

The nutrition plan below is built around these two risks.

## The core principles

### 1. A protein floor, set in grams, not percent

Studies and practice consistently point to roughly **1.6 g of protein per kg of bodyweight per day** as the working floor on GLP-1 therapy. Some clinicians push the floor to 1.8 or 2.0 g/kg for older adults or those doing resistance training.

A grams-based floor is harder to undercut than a percentage on days where total intake crashes. Use the [protein calculator](/blog/protein-calculator) for your number and the GLP-1 toggle.

### 2. Small, frequent meals

Because gastric emptying is slowed, a 700 kcal dinner that felt normal pre-medication can sit heavy and trigger nausea. Three small meals plus one or two protein-forward snacks is usually better tolerated than two big meals.

### 3. Hydration and fiber

Constipation is one of the most common reasons people stop GLP-1 therapy early. Two cheap fixes:

- Aim for 30–35 ml of water per kg of bodyweight per day. For a 70 kg adult that's roughly 2.1–2.5 L.
- Aim for 25–35 g of fiber per day from vegetables, legumes, whole grains, and fruit.

### 4. Watch the early-titration triggers

During the first 8–16 weeks (and at each dose escalation), the foods most commonly reported as triggering nausea, reflux, or stomach pain include:

- High-fat fried foods (fried chicken, fries, deep-fried snacks)
- Very greasy or oily meals (rich curries, heavy cream sauces)
- Large portions of red meat in a single sitting
- Carbonated drinks with meals
- Strong alcohol on an empty stomach
- Very spicy dishes during the first few weeks

Many people who can't tolerate these foods at week 4 are fine with them by month 4 — the trigger list is dose-dependent and adaptive.

### 5. Plate composition that works

For each main meal, the structure clinicians often suggest is:

- **One-third lean protein** (chicken, fish, tofu, lean beef, Greek yogurt, eggs)
- **One-third high-fiber vegetables** (leafy greens, cruciferous, peppers, mushrooms)
- **One-third complex carbs** (oats, quinoa, brown rice, sweet potato, legumes)
- A small amount of healthy fat (olive oil, avocado, nuts)

This is consistent with the Mediterranean and DASH patterns that have decades of cardiovascular and metabolic research behind them.

## Drug-by-drug: food-relevant differences

This table covers nutrition-relevant differences only. For medication mechanism, dosing, contraindications, and side-effect management, refer to the FDA prescribing information for each drug and your clinician.

| Feature | Ozempic | Wegovy | Mounjaro | Zepbound |
|---|---|---|---|---|
| Molecule | Semaglutide | Semaglutide | Tirzepatide (GIP+GLP-1) | Tirzepatide (GIP+GLP-1) |
| Approved use | Type 2 diabetes | Chronic weight management | Type 2 diabetes | Chronic weight management |
| Appetite suppression intensity | Strong | Strong (higher doses) | Very strong | Very strong |
| Typical GI side effects | Nausea, constipation | Nausea, constipation | Nausea, diarrhea | Nausea, diarrhea |
| Reported sensitivity to fatty meals | High during titration | High during titration | High during titration | High during titration |
| Reported sensitivity to alcohol | Reduced tolerance | Reduced tolerance | Reduced tolerance | Reduced tolerance |

_Nutrition-relevant differences between the major GLP-1 medications. Generalizations — your prescriber will personalize._

The practical nutrition advice — protein floor, small frequent meals, hydration, fiber, limited fried food — is the same across all four. Differences are mostly in the GI side-effect profile, which affects which foods you experiment with first.

## A sample 7-day structure (not a prescription)

This is a structural example showing how the principles above translate into meals. Portion sizes assume a 70 kg adult targeting ~1,500–1,800 kcal/day with ~120 g protein. Scale to your own number. Talk to a registered dietitian before adopting it as a long-term plan.

### Day 1
- **Breakfast:** Greek yogurt (200 g) + berries + 1 tbsp ground flaxseed + 2 tbsp almonds (~30 g protein, ~350 kcal)
- **Lunch:** Grilled chicken salad — 120 g chicken, mixed greens, chickpeas, olive oil + lemon (~35 g protein, ~450 kcal)
- **Snack:** Cottage cheese (150 g) + apple (~20 g protein, ~200 kcal)
- **Dinner:** Baked salmon (120 g) + quinoa (100 g cooked) + roasted broccoli (~35 g protein, ~500 kcal)

### Day 2
- **Breakfast:** Two-egg veggie omelette + whole-grain toast + avocado (~22 g protein, ~400 kcal)
- **Lunch:** Lentil soup + side of grilled chicken (80 g) (~30 g protein, ~450 kcal)
- **Snack:** Whey or pea-protein shake in water (~25 g protein, ~120 kcal)
- **Dinner:** Tofu and vegetable stir-fry with brown rice (~30 g protein, ~500 kcal)

### Day 3
- **Breakfast:** Overnight oats with Greek yogurt, chia seeds, banana (~25 g protein, ~400 kcal)
- **Lunch:** Turkey & hummus wrap on whole-wheat tortilla, side salad (~35 g protein, ~500 kcal)
- **Snack:** Hard-boiled eggs (2) + cherry tomatoes (~13 g protein, ~160 kcal)
- **Dinner:** Lean beef chili with kidney beans + side of greens (~35 g protein, ~500 kcal)

### Day 4
- **Breakfast:** Protein smoothie (whey, frozen berries, spinach, almond milk, 1 tbsp peanut butter) (~30 g protein, ~380 kcal)
- **Lunch:** Tuna salad on mixed greens with chickpeas (~35 g protein, ~450 kcal)
- **Snack:** Greek yogurt + walnuts (~17 g protein, ~250 kcal)
- **Dinner:** Grilled chicken thigh (skin off, 120 g) + roasted sweet potato + asparagus (~32 g protein, ~500 kcal)

### Day 5
- **Breakfast:** Cottage cheese (150 g) + pineapple + 2 tbsp chia seeds (~20 g protein, ~300 kcal)
- **Lunch:** Egg-white scramble with smoked salmon + sourdough toast (~35 g protein, ~450 kcal)
- **Snack:** Edamame (1 cup) + small handful almonds (~20 g protein, ~250 kcal)
- **Dinner:** White-fish curry (mild) + brown rice + cucumber raita (~30 g protein, ~500 kcal)

### Day 6
- **Breakfast:** Two scrambled eggs + Greek yogurt + berries (~25 g protein, ~350 kcal)
- **Lunch:** Chickpea & quinoa bowl with feta and olive oil (~25 g protein, ~500 kcal)
- **Snack:** Skyr or 0% Greek yogurt + cinnamon (~17 g protein, ~150 kcal)
- **Dinner:** Lean ground turkey lettuce wraps + brown rice on the side (~40 g protein, ~500 kcal)

### Day 7 (flexible / dinner-out friendly)
- **Breakfast:** Protein oatmeal — oats cooked with milk + whey + walnuts + banana (~30 g protein, ~450 kcal)
- **Lunch:** Light Mediterranean plate — grilled fish, hummus, cucumber, tomato, small pita (~30 g protein, ~450 kcal)
- **Snack:** Protein shake or Greek yogurt (~25 g protein, ~150 kcal)
- **Dinner (restaurant or home):** Grilled chicken or fish + double vegetables + small starch — skip the bread basket and the fried appetizer (~35 g protein, ~550 kcal)

## What to do if you can't finish meals

This happens — appetite on a GLP-1 can collapse without warning. The published guidance, and what most dietitians recommend, is:

- **Eat protein first.** If you can only get a few bites in, make them the protein on the plate.
- **Use liquid protein.** A whey or pea-protein shake delivers 25–30 g in 250–300 ml — much easier than chewing.
- **Don't skip — shrink.** A 200 kcal mini-meal of yogurt and berries protects your weekly pattern better than skipping entirely.
- **Talk to your prescriber.** Persistent inability to finish meals, weight loss faster than 1% per week, or any sign of dehydration warrants a clinical conversation.

## When to talk to your prescriber

The lines below are nutrition-side flags. They are not exhaustive medical criteria.

- Persistent nausea or vomiting beyond the first two weeks of a new dose
- Inability to drink at least 1.5 L of fluid per day
- Weight loss faster than 1% of bodyweight per week, sustained
- Significant muscle weakness, dizziness, or fatigue
- Constipation lasting more than 3 days despite hydration and fiber
- Any new abdominal pain that is severe or persistent

## Tracking the plan

Most calorie trackers were not designed for the GLP-1 use case — where the user eats *less* than the floor and the app keeps suggesting "eat more." Callie's daily plan is BMI-matched and adjusts to your actual intake pattern, and the [protein calculator](/blog/protein-calculator) has a GLP-1 toggle that raises the floor automatically. Pair that with photo or voice logging and you can hit your protein floor in 30 seconds per meal.

## Primary sources and further reading

- FDA Prescribing Information — Ozempic (semaglutide injection), Novo Nordisk
- FDA Prescribing Information — Wegovy (semaglutide injection), Novo Nordisk
- FDA Prescribing Information — Mounjaro (tirzepatide injection), Eli Lilly
- FDA Prescribing Information — Zepbound (tirzepatide injection), Eli Lilly
- STEP clinical trial program (semaglutide for chronic weight management), published in The New England Journal of Medicine
- SURMOUNT clinical trial program (tirzepatide for chronic weight management), published in The New England Journal of Medicine
- AACE 2023 Obesity Clinical Practice Guideline (American Association of Clinical Endocrinology)
- The Obesity Society — clinical practice statements on pharmacotherapy and nutrition

Always confirm any change to your eating pattern with the clinician managing your GLP-1 prescription. This article is a summary of published guidance and common clinical recommendations; it cannot personalize to your medications, conditions, or goals.

---

### Macro Calculator with Cheat Day Adjuster (2026)

URL: https://www.mycallie.app/en/blog/macro-calculator-with-cheat-day
Markdown: https://www.mycallie.app/md/en/blog/macro-calculator-with-cheat-day.md
Published: 2026-05-27 · Updated: 2026-05-27
Tags: macro calculator, cheat day calculator, weekly calorie average, diet refeed calculator

If you came here from Google, you want the math.

## The 30-second answer

A diet works on the **weekly calorie average**, not the daily number. So if you eat 3,200 kcal at Saturday brunch on a target of 1,800 kcal/day, you don't need to undo it — you redistribute the surplus across the rest of the week.

**Formula:** `(daily target × 7) − cheat-day kcal) ÷ remaining days = new daily target`

Or just use this:

[Interactive macro calculator with cheat-day adjuster — use it on the page.]

## Why weekly averaging is the right unit

Body weight is set by energy balance over weeks, not days. Three things make daily numbers misleading:

1. **Water swings dwarf real weight change.** A high-sodium meal can add 1–1.5 kg of water overnight. That's not fat.
2. **Glycogen replenishment.** Each gram of glycogen binds ~3 g of water. A carb-heavy day adds 1–2 kg of glycogen + water that's gone within a few days.
3. **Measurement noise.** Daily weigh-ins have variance of ±1–2% just from food, fluid, and bathroom timing.

So the right question is not "what did I eat today?" but **"what was my 7-day average?"** If the average is at target, you're losing weight on schedule even if individual days look erratic.

## The cheat-day math, worked

Say you're a 70 kg adult cutting at 1,800 kcal/day:

- **Weekly target:** 1,800 × 7 = 12,600 kcal
- **Saturday cheat:** 3,500 kcal (brunch out + drinks + dessert)
- **Surplus from cheat day:** 3,500 − 1,800 = 1,700 kcal
- **Remaining 6 days:** 12,600 − 3,500 = 9,100 kcal
- **Daily target for those 6 days:** 9,100 ÷ 6 = **1,517 kcal**

So you drop ~280 kcal from your normal target for six days. That's noticeable but not punitive — pull a snack, lighten one meal.

## When the math doesn't work

If your cheat day is so big that the recovery days would drop below 1,200 kcal, you've over-shot the absorbable range:

- **3,500 kcal cheat day, 7-day window:** recovery days = ~1,517 — fine
- **5,000 kcal cheat day, 7-day window:** recovery days = ~1,267 — borderline
- **5,000 kcal cheat day, 3-day window:** recovery days = ~700 — unsafe; widen the window or accept the week runs over

A floor of 1,200 kcal for women / 1,500 kcal for men is the practical safety line. Below that you get fatigue, poor recovery, and binge-rebound risk. If the math wants you below the floor, the honest answer is: this week will run over your target by a few hundred kcal/day. Take the small setback over an unsafe cut.

## How macros redistribute alongside the calories

When you cut calories on recovery days, the simplest split is to:

- **Keep protein constant** (in grams, not percent). Protein is the macro that protects muscle in a deficit; don't sacrifice it.
- **Pull the surplus from carbs and fat proportionally.**

Example — Recovery-day target 1,517 kcal at 30/40/30 (C/P/F):

| Macro | Calories | Grams |
|---|---|---|
| Carbs (30%) | 455 | 114 g |
| Protein (40%) | 607 | 152 g |
| Fat (30%) | 455 | 51 g |

The calculator above does this math for you. For more on hitting protein specifically, see the [protein calculator](/blog/protein-calculator).

## Cheat day vs refeed day — different tools

These get used interchangeably, but they're not the same:

| | Cheat day | Refeed day |
|---|---|---|
| Purpose | Adherence, social, mental break | Hormone and performance support |
| Calories | Well above target — uncontrolled | At maintenance (TDEE) |
| Macros | Whatever | Higher carbs specifically |
| Frequency | As needed | Every 1–2 weeks during long cuts |
| Foods | Anything | Usually whole-food, dense carbs |

A refeed is the more research-backed tool — multiple studies suggest planned maintenance breaks during long cuts help with leptin, thyroid output, and training capacity. A cheat day is the adherence tool — it's what keeps the diet psychologically sustainable.

You can use both. Many lifters do a Saturday cheat day (mental) and a Wednesday refeed (physiological).

## How to make cheat days "not ruin the streak"

This is exactly what Callie's [streak-preserving cheat day](/) feature does in the app: schedule the day in advance, the app auto-redistributes the surplus across the next several days, and your streak stays intact. You can do the same math manually with the calculator above — the point is that the cheat day becomes a planned event with a recovery plan, not a derailment.

Three habits make planned cheat days work:

1. **Schedule it.** Mark the day in advance. Spontaneous overeating is what makes "cheat days" turn into cheat weeks.
2. **Plan the next 3–7 days.** Before the cheat day, decide the redistribution window. Write the recovery-day target somewhere visible.
3. **Don't double-down.** A common mistake: after a big day, people decide to "really cut" and drop to 1,000 kcal as punishment. That triggers binge-rebound. Use the math, not guilt.

## Special case: weekend cheat days

Two cheat days back-to-back (Friday + Saturday) compound fast. If you eat 3,000 kcal both days on an 1,800 target:

- **Combined surplus:** 2 × (3,000 − 1,800) = 2,400 kcal
- **Recovery days (Mon–Thu, 4 days):** weekly target 12,600 − 6,000 = 6,600 → 1,650/day

That's a sharper drop. Many people find it easier to take one cheat day per week than to spread the indulgence across two — the recovery math is gentler.

## What "cheat" actually means in nutrition

There's no morally bad food. The word "cheat" is just shorthand for "above-target day." Some practitioners prefer "free day" or "flex day" because the framing matters — the goal is calm, planned variability, not punishment-and-reward.

The math is the same regardless of what you call it.

## When this approach falls apart

Weekly averaging works when:

- You're tracking honestly (under-logging defeats the math — see [why a calorie deficit might not be working](/blog/calorie-deficit-not-losing-weight))
- Cheat days are occasional, not most days
- You're not using "I'll recover next week" as a permanent excuse

If you find yourself in a constant cycle of cheat day → "recovery" → another cheat day before the recovery completes, the diet target is probably too aggressive. Step the deficit down to 10–15% and aim for boring consistency instead of weekly drama.

## Putting it all together

1. Pick your daily target from your TDEE.
2. Set your protein floor in grams (use the [protein calculator](/blog/protein-calculator)).
3. When a cheat day happens, use the calculator above to compute the recovery-day target.
4. Adjust carbs and fat downward on recovery days; keep protein constant.
5. Track the weekly average, not the daily number.

The weekly average is the only number that matters. A single high day is recoverable; a permanent 200 kcal/day excess that nobody's tracking is not. Plan the cheat, plan the recovery, then forget about it.

---

### Protein Calculator: How Much Protein Do You Actually Need? (2026)

URL: https://www.mycallie.app/en/blog/protein-calculator
Markdown: https://www.mycallie.app/md/en/blog/protein-calculator.md
Published: 2026-05-27 · Updated: 2026-05-27
Tags: protein calculator, how much protein do I need, protein per kg bodyweight, protein on Ozempic

If you came here from Google, you want a number. Here it is.

## The 30-second answer

**Take your bodyweight in kg and multiply by 1.2 to 1.6.** That's your daily protein target in grams. If you're cutting, on a GLP-1, or trying to build muscle, multiply by 1.6 to 2.2 instead.

Skip the math:

[Interactive protein calculator — use it on the page.]

## Why the RDA (0.8 g/kg) is too low for most adults

The US RDA for protein — 0.8 g/kg/day — is the amount needed to prevent deficiency in sedentary adults. It is not the amount needed to optimize body composition, recovery, or healthy aging.

A 2022 ISSN position stand and decades of follow-up research consistently land on these working numbers for active adults:

| Goal | Protein (g per kg bodyweight) | Protein (g per lb) |
|---|---|---|
| Sedentary maintenance (RDA floor) | 0.8 | 0.36 |
| General health, light activity | 1.2–1.4 | 0.55–0.64 |
| Recreational lifting, weight maintenance | 1.4–1.8 | 0.64–0.82 |
| Cutting (calorie deficit) | 1.6–2.4 | 0.73–1.09 |
| Muscle gain (calorie surplus) | 1.6–2.2 | 0.73–1.00 |
| On a GLP-1 (Ozempic / Wegovy / Mounjaro) | 1.6–2.0 | 0.73–0.91 |
| Hard gainer (struggling to gain) | 2.2–3.4 | 1.00–1.55 |

## Daily protein targets — quick reference

Midpoint targets in grams per day, by bodyweight and goal.

### Maintenance (1.4 g/kg)

| Bodyweight | Daily protein |
|---|---|
| 50 kg / 110 lb | 70 g |
| 60 kg / 132 lb | 84 g |
| 70 kg / 154 lb | 98 g |
| 80 kg / 176 lb | 112 g |
| 90 kg / 198 lb | 126 g |
| 100 kg / 220 lb | 140 g |

### Cutting or GLP-1 (2.0 g/kg)

| Bodyweight | Daily protein |
|---|---|
| 50 kg / 110 lb | 100 g |
| 60 kg / 132 lb | 120 g |
| 70 kg / 154 lb | 140 g |
| 80 kg / 176 lb | 160 g |
| 90 kg / 198 lb | 180 g |
| 100 kg / 220 lb | 200 g |

### Hard gainer (2.8 g/kg)

| Bodyweight | Daily protein |
|---|---|
| 50 kg / 110 lb | 140 g |
| 60 kg / 132 lb | 168 g |
| 70 kg / 154 lb | 196 g |
| 80 kg / 176 lb | 224 g |
| 90 kg / 198 lb | 252 g |

## The per-meal split most people get wrong

Daily total matters most, but distribution is a real second-order lever. Research on muscle protein synthesis (MPS) suggests:

- The MPS response to a single meal plateaus around **30–40 g** of high-quality protein for most adults. Older adults need the higher end because of anabolic resistance.
- Going much higher in a single meal (say, 80 g at dinner) is not "wasted" — the extra protein is still used for general metabolism — but it doesn't stack additional muscle-building benefit on that meal.
- Splitting your daily target into 3–4 meals of 30–40 g each is the practical sweet spot.

A 70 kg adult eating 140 g/day:

| Meals/day | Per meal |
|---|---|
| 3 meals | ~47 g (a little above the MPS ceiling — fine) |
| 4 meals | ~35 g (right in the sweet spot) |
| 5 meals (snacks count) | ~28 g (slightly under — still works) |
| 1 meal (OMAD) | 140 g (a lot at once; OMAD is not optimal for muscle, but works for weight management) |

## Special case: protein on a GLP-1 medication

GLP-1 drugs (Ozempic, Wegovy, Mounjaro, Zepbound) suppress appetite so strongly that most users involuntarily drop total food intake by 20–35%. That's good for weight loss but bad for muscle if protein drops proportionally.

- **Set a 1.6 g/kg floor, not a percentage of calories.** A grams-based floor is harder to undercut on low-appetite days.
- **Front-load protein.** Many GLP-1 users eat well at breakfast and lunch but lose interest by dinner. Get 50–60% of your daily protein in by 2 pm.
- **Liquid protein when solid food won't go down.** A whey or pea-protein shake delivers 25–30 g in 250–300 ml.

For the full GLP-1 nutrition framework, see [GLP-1 diet plan: what to eat on Ozempic, Wegovy & Mounjaro](/blog/glp-1-diet-plan-ozempic-wegovy-mounjaro).

For the lean-mass side specifically — what the 2025 four-society advisory recommends and why per-meal distribution is the real constraint — see [GLP-1 and muscle loss](/blog/glp-1-muscle-loss-protein).

## Special case: the hard gainer

If you're consistently trying to gain weight and lifting hard, the upper end of the protein range (2.2–3.4 g/kg) helps drive lean-mass accrual specifically rather than just total weight. Combine with a calorie surplus of 250–500 kcal/day. The full playbook is in our [hard gainer guide](/blog/healthy-weight-gain-guide).

## How to actually hit the number

The protein gap most adults face is not strategy, it's tracking. People reliably under-log meat portions and forget to count protein in carby foods (bread has 3–4 g per slice; pasta has 7–8 g per cooked cup). Two compounding errors:

1. Under-logging protein on photos (chicken portions look smaller than they are).
2. Forgetting incidental protein in side dishes.

The fix is a tracker that recognizes whole meals (not single items) and lets you adjust portion size after the fact. Callie's photo + voice logging is built for this — and the [review-before-confirm step](/blog/how-to-track-calories-from-a-photo) catches portion misjudgments before they bake into your weekly average.

## Common protein-rich foods, sorted by g per 100 kcal

When you're in a calorie deficit, density matters. These foods give you the most protein per calorie spent:

| Food | Protein per 100 kcal |
|---|---|
| Egg whites | ~21 g |
| Whey isolate | ~22 g |
| Cod, plain | ~21 g |
| Chicken breast, skinless | ~18 g |
| Greek yogurt 0% | ~17 g |
| Tofu, firm | ~12 g |
| Lentils, cooked | ~7 g |
| Whole eggs | ~8 g |
| Almonds | ~3.5 g |

## When to recalculate

Your protein target follows your bodyweight, so every 5 kg / 10 lb of weight change is worth a recheck. Also recheck if you:

- Start or stop a GLP-1 medication
- Switch from maintenance to a cut or surplus
- Start strength training after a sedentary period
- Cross age 65 (older adults benefit from the upper end of the range due to anabolic resistance)

Bodyweight in kg × 1.6 is a defensible target for almost any active adult. If you're cutting or on a GLP-1, push it to 2.0. Hit that number 5–6 days a week — the result compounds far more than perfect macro hair-splitting on any single day.

---

### Best AI Calorie Tracker Apps of 2026 (Honest Ranking)

URL: https://www.mycallie.app/en/blog/best-ai-calorie-tracker-apps-2026
Markdown: https://www.mycallie.app/md/en/blog/best-ai-calorie-tracker-apps-2026.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: best AI calorie tracker, best calorie counter app, AI nutrition apps, calorie tracker comparison

Disclosure first, because it matters: **Callie is our app.** This article is an honest ranking, but you're right to assume bias. To make the bias inspectable, we've published our scoring rubric *before* the rankings — so you can decide if the categories and weights are fair, and recalculate with your own weights if you disagree.

If we wanted to fake a "best of" list, we'd skip the rubric and just put Callie at #1. Read the rubric, then judge.

## The rubric (published *before* scoring)

Each app is scored 0–10 across six categories. The total is a weighted sum.

| Category | Weight | What it measures |
|---|---|---|
| **Logging speed** | 25% | Median seconds per meal for a typical user, across the app's primary input modalities |
| **Accuracy** | 20% | Mean absolute error vs kitchen-scale, on a standard 20-meal benchmark |
| **Language support** | 15% | Number of fully localized UI languages and quality of non-English AI chat |
| **Free-tier breadth** | 15% | What's available without paying |
| **AI coaching** | 15% | Quality of conversational guidance, plateau detection, meal-timing patterns |
| **Ecosystem / integrations** | 10% | Wearable + health-kit integrations, web access, export options |

Total weighted score out of 10.

## Apps included

We evaluated apps available on both iOS and Android with at least 100k installs that explicitly market AI calorie tracking or are dominant in the category:

- Callie (our app)
- MyFitnessPal
- Cal AI
- Lose It!
- Noom
- Yazio

Apps excluded: Cronometer (excellent app but not primarily AI-first), Lifesum (legacy positioning), niche regional apps not available in the US App Store.

## The rankings (with caveats)

| Rank | App | Score | Best for |
|---|---|---|---|
| 1 | Callie | 8.3/10 | Multi-modal AI logging, multi-language users, AI coach |
| 2 | MyFitnessPal | 7.4/10 | Largest food database, mature ecosystem, branded-food barcode users |
| 3 | Cal AI | 7.0/10 | Single-purpose photo-only logging, clean UI |
| 4 | Yazio | 6.5/10 | Strong recipe library, intermittent fasting integration |
| 5 | Lose It! | 6.4/10 | Established alternative to MyFitnessPal, good barcode scanner |
| 6 | Noom | 5.9/10 | Behavioral coaching (note: this is really a coaching app with tracking attached) |

Ranks 2 and 3 are now the same company: MyFitnessPal acquired Cal AI in a deal that closed December 2025 and was announced March 2, 2026, and Cal AI now runs on MyFitnessPal's food database. Cal AI remains a standalone app with its own subscription. See [MyFitnessPal vs Cal AI](/blog/myfitnesspal-vs-cal-ai) for what changed. Note also that Lose It! (rank 5) moved its barcode scanner, Snap It photo logging, and macro tracking into Premium — details in [MyFitnessPal vs Lose It!](/blog/myfitnesspal-vs-lose-it).

### Category breakdown

| Category | Callie | MyFitnessPal | Cal AI | Yazio | Lose It! | Noom |
|---|---|---|---|---|---|---|
| Logging speed | 9 | 5 | 8 | 6 | 5 | 4 |
| Accuracy | 8 | 8 | 7 | 7 | 7 | 6 |
| Language support | 9 | 5 | 4 | 6 | 5 | 4 |
| Free-tier breadth | 9 | 5 | 7 | 6 | 7 | 3 |
| AI coaching | 9 | 4 | 5 | 5 | 4 | 8 |
| Ecosystem | 6 | 9 | 5 | 6 | 7 | 6 |

A few things this honestly shows:

- **Callie loses ecosystem.** MyFitnessPal's third-party integrations and history are 10 years deep. We're not there yet.
- **Noom's coaching is genuinely strong** — that's their core competency. We score Noom lower overall because as a *calorie tracker*, it underdelivers on speed and database.
- **Cal AI's UI is excellent.** The 7/10 logging-speed score doesn't capture how *focused* the experience is.

## When each app is right

**Choose Callie if** you want fast multi-modal logging (photo/voice/text) with **review-and-edit before logging** (no surprise auto-logs), a **daily diet plan matched to your BMI** with meal and exercise suggestions, a **GitHub-style streak dashboard**, **scheduled cheat days that protect your streak**, a localized UI (English, French, Spanish, German, Arabic, Japanese) with an AI coach in any language, all in the free tier.

**Choose MyFitnessPal if** you log mostly packaged branded foods by barcode, you've used it for years and your workflow is built around it, or you need specific third-party integrations.

**Choose Cal AI if** you want a streamlined photo-only experience and don't need voice, text, or coaching.

**Choose Yazio if** you want strong recipe integration and IF support in a single app.

**Choose Lose It! if** you want a MyFitnessPal-style experience with a slightly cleaner UI.

**Choose Noom if** you want structured behavioral coaching and are willing to pay subscription fees — and treat the tracker as secondary.

## Methodology notes

- **Accuracy benchmark:** 20 plated meals weighed on a 0.1g kitchen scale; each photographed with a fork in frame and logged once per app; compared to USDA FoodData Central calorie densities. Methodology details and per-meal data available in our [Callie vs Cal AI comparison](/compare/callie-vs-cal-ai).
- **Speed benchmark:** Median end-to-end time per meal across the app's primary modalities, by a moderate-experience user.
- **Tests conducted:** April–May 2026. App versions tested as of May 19, 2026.

## What we got wrong in last year's list (we update this annually)

In a 2025 internal draft we ranked Callie above MyFitnessPal partly because MyFitnessPal had a poorer iOS launch experience. They've improved it. The current rubric reflects that fairly — MyFitnessPal lost 1 point in our previous "Onboarding" sub-category and we re-weighted "Logging speed" higher because it correlates more strongly with adherence.

We'll update this list in May 2027 and document changes again.

## A word on "best of" lists

Most "best calorie tracker" lists on the web are affiliate revenue strategies, ranked by which app pays the highest commission. We're disclosing the bias because: (a) you should know, (b) the alternative is pretending we don't have one, (c) the rubric makes the ranking falsifiable. If you disagree with our weights, you can recompute.

## Related reading

- [The Complete Guide to AI Calorie Tracking](/guides/ai-calorie-tracking) — primer on the underlying technology.
- [Callie vs MyFitnessPal](/compare/callie-vs-myfitnesspal) — detailed head-to-head.
- [Callie vs Cal AI](/compare/callie-vs-cal-ai) — detailed head-to-head with the closest competitor.
- [Callie vs Lose It!](/compare/callie-vs-lose-it) — free tier vs Premium, and Lose It!'s lifetime purchase option.
- [MyFitnessPal vs Cal AI](/blog/myfitnesspal-vs-cal-ai) — the 2026 acquisition, and how the two apps differ now that they share an owner.
- [MyFitnessPal vs Lose It!](/blog/myfitnesspal-vs-lose-it) — what each app actually gives you for free.

## Sources

1. Internal Callie 20-meal accuracy benchmark (May 2026). See [Callie vs Cal AI](/compare/callie-vs-cal-ai) for methodology.
2. App Store and Google Play listings for all 6 apps reviewed (May 2026).
3. USDA FoodData Central. https://fdc.nal.usda.gov/

---

### Healthy Weight Gain: A Guide for Hard Gainers (2026)

URL: https://www.mycallie.app/en/blog/healthy-weight-gain-guide
Markdown: https://www.mycallie.app/md/en/blog/healthy-weight-gain-guide.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: gain weight app, weight gain, healthy weight gain, calorie surplus

Most calorie tracking advice is written for people who want to lose weight. If you're a "hard gainer" — high metabolism, low appetite, slow muscle progress — most of that advice is exactly backwards.

This guide is the gain-side equivalent. Math, food, training, and how to actually pull off a clean bulk.

## The calorie surplus math

Eat above TDEE; gain weight. The classic estimate is 3,500 extra kcal ≈ 1 lb (0.45 kg) of body mass. So a 500 kcal/day surplus produces ~1 lb/week of gain.

The catch: not all of that lb is muscle. The split between muscle and fat depends on:

1. **Your training stimulus** — without progressive strength training, the surplus mostly becomes fat.
2. **Surplus size** — bigger surpluses past a point add fat, not muscle.
3. **Protein intake** — too low and you can't build muscle even with surplus + training.
4. **Training experience** — beginners gain muscle much faster than advanced lifters.

## Surplus size recommendations

| Surplus | % above TDEE | Weekly gain | Muscle-to-fat ratio (with training) |
|---|---|---|---|
| Lean bulk | 5–10% | 0.1–0.25 kg | ~70/30 |
| Standard bulk | 10–20% | 0.25–0.5 kg | ~50/50 |
| Dirty bulk | 20–40% | 0.5–1.0 kg | ~30/70 |

For most lifters, **standard bulk (10–15% surplus, ~0.25 kg/week)** is the sweet spot. A clean lean bulk preserves a visible physique year-round; a dirty bulk requires a longer cut afterward and erodes most of the strength advantage.

## Protein target

The single most important macro for muscle gain. The research consistently supports **1.6–2.2 g/kg body weight per day** ([Morton et al., 2018](https://bjsm.bmj.com/content/52/6/376) meta-analysis of 49 studies).

For a 70 kg (155 lb) adult, that's 112–155 g protein/day. Practical sources:

| Protein source | Protein per 100g | Kcal per 100g |
|---|---|---|
| Chicken breast (cooked) | 31g | 165 |
| Whey protein powder | 80g | 380 |
| Salmon (cooked) | 25g | 208 |
| Lean ground beef (90/10) | 27g | 217 |
| Greek yogurt (full-fat) | 9g | 130 |
| Cottage cheese (whole milk) | 11g | 98 |
| Eggs (whole) | 13g | 155 |
| Tofu (firm) | 17g | 144 |
| Lentils (cooked) | 9g | 116 |
| Tempeh | 19g | 192 |

## Calorie-dense foods (gain without overstuffing)

For hard gainers, the hardest part is hitting calories before getting full. Eat foods with high kcal per gram:

| Food | Kcal per 100g | Protein per 100g | Notes |
|---|---|---|---|
| Olive oil | 884 | 0 | Add 1 tbsp to dinner = +120 kcal |
| Almonds | 579 | 21g | Snack — 30g handful = +175 kcal |
| Peanut butter | 588 | 25g | 2 tbsp = +190 kcal |
| Avocado | 160 | 2g | One = +250 kcal, healthy fats |
| Granola | 489 | 10g | Adds up fast in milk |
| Dried oats | 379 | 13g | With whole milk + honey = easy 500-kcal breakfast |
| Whole milk | 61 | 3.2g | Switch from skim = +1,500 kcal/week for a 3-glass/day drinker |
| Dark chocolate (85%) | 598 | 8g | A square is ~30 kcal |
| Trail mix | ~480 | ~14g | High-density snack |
| Whole-grain pasta + olive oil | ~400 (cooked + oil) | ~10g | Carb-heavy lunch |

**The fastest hard-gainer hack:** a daily smoothie. Whole milk (250 kcal) + banana (105) + 2 tbsp peanut butter (190) + whey scoop (120) + oats (150) = 815 kcal in something you drink in 90 seconds. Hard-gainer appetite throttles taste, not liquid volume.

## Sample bulking day (target ~3,000 kcal, 150g protein)

| Meal | Foods | Kcal | Protein |
|---|---|---|---|
| Breakfast | Oats (80g) + whole milk (300ml) + banana + 2 tbsp peanut butter | 680 | 23g |
| Snack | Greek yogurt + honey + 30g almonds | 380 | 17g |
| Lunch | 200g chicken thigh + 1.5 cups cooked rice + 1 tbsp olive oil + veg | 760 | 50g |
| Pre-workout snack | Banana + protein shake | 230 | 25g |
| Dinner | 200g salmon + sweet potato + asparagus + olive oil | 700 | 40g |
| Evening | Cottage cheese + berries | 200 | 20g |
| **Total** | | **2,950** | **175g** |

## Training matters more than the food

A surplus without training builds fat. A surplus *with* progressive strength training is the lever that decides where the calories go.

The minimum effective dose:

- **3–4 strength sessions/week.**
- **Compound lifts:** squat, deadlift, bench, row, overhead press, pull-up.
- **Progressive overload:** add weight, reps, or sets every 1–2 weeks.
- **8–20 sets per muscle group per week** (more than most people do).

You don't have to overcomplicate it. Show up consistently, lift heavier than last week, eat enough protein.

## When you're not gaining

If you've been at your target surplus for 3 weeks with no weight movement:

1. **Re-log a typical day** and verify you're actually hitting the calorie target. Hard gainers chronically under-eat what they planned.
2. **Add 200 kcal/day** via liquid calories (a smoothie, whole milk, juice).
3. **Check NEAT.** Some people unconsciously fidget more on a surplus — set a step ceiling (~10,000) rather than a floor.
4. **Sleep.** Under 7 hours nightly cuts muscle protein synthesis by ~20% and increases ghrelin.

## How Callie supports bulking

Set Callie's goal to "gain weight." The target adjusts to TDEE + your chosen surplus. The AI coach flags low-protein days and low-total-calorie days — the two failure modes for hard gainers. Voice logging in particular helps because you'll be eating frequently and don't want to type 6 times a day.

## Related reading

- [Calorie Deficit Calculator & Complete Guide](/guides/calorie-calculator) — TDEE math (works for surpluses too).
- [How Many Calories Should I Eat to Lose Weight?](/blog/how-many-calories-to-lose-weight) — the deficit side.
- [The Complete Guide to AI Calorie Tracking](/guides/ai-calorie-tracking) — primer on the underlying tech.

## Sources

1. Morton RW, et al. (2018). "A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults." *Br J Sports Med* 52(6):376-384. https://bjsm.bmj.com/content/52/6/376
2. Helms ER, et al. (2014). "Evidence-based recommendations for natural bodybuilding contest preparation: nutrition and supplementation." *J Int Soc Sports Nutr* 11:20.
3. Aragon AA, Schoenfeld BJ. (2020). "Magnitude and Composition of the Energy Surplus for Maximizing Muscle Hypertrophy." *Strength and Conditioning Journal* 42(5):79-86.
4. USDA FoodData Central. https://fdc.nal.usda.gov/
5. Schoenfeld BJ, Aragon AA. (2018). "How much protein can the body use in a single meal for muscle-building? Implications for daily protein distribution." *J Int Soc Sports Nutr* 15:10.

> **Important:** Educational only. If you have a history of disordered eating, a thyroid condition, GI issues, or a clinical reason for low body weight, work with a registered dietitian.

---

### How Many Calories Should I Eat to Lose Weight? (2026)

URL: https://www.mycallie.app/en/blog/how-many-calories-to-lose-weight
Markdown: https://www.mycallie.app/md/en/blog/how-many-calories-to-lose-weight.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: how many calories to lose weight, daily calorie intake, weight loss, calorie deficit

If you came here from Google, you want a number and a quick reason. Here it is.

## The 30-second answer

**Eat 15–25% below your TDEE.** For most adults that's a daily target between 1,400 and 2,200 kcal, depending on your size and activity. Use the [calorie deficit calculator](/guides/calorie-calculator) to get your exact number.

If you'd rather skim a table than do math:

## Daily calorie targets — quick reference

These assume a 20% deficit (recommended starting point) and use Mifflin-St Jeor BMR plus a moderate-activity multiplier (1.55).

### Women, moderate activity

| Age | Weight (kg) | Height (cm) | Daily target (kcal) |
|---|---|---|---|
| 25 | 55 | 160 | 1,510 |
| 25 | 65 | 165 | 1,640 |
| 25 | 75 | 170 | 1,770 |
| 35 | 60 | 162 | 1,540 |
| 35 | 70 | 167 | 1,670 |
| 35 | 80 | 172 | 1,800 |
| 45 | 65 | 160 | 1,550 |
| 45 | 75 | 165 | 1,680 |
| 45 | 85 | 170 | 1,810 |
| 55 | 70 | 162 | 1,560 |
| 55 | 80 | 167 | 1,690 |

### Men, moderate activity

| Age | Weight (kg) | Height (cm) | Daily target (kcal) |
|---|---|---|---|
| 25 | 70 | 175 | 2,140 |
| 25 | 80 | 180 | 2,300 |
| 25 | 90 | 185 | 2,440 |
| 35 | 75 | 175 | 2,130 |
| 35 | 85 | 180 | 2,290 |
| 35 | 95 | 185 | 2,430 |
| 45 | 80 | 175 | 2,110 |
| 45 | 90 | 180 | 2,270 |
| 45 | 100 | 185 | 2,410 |
| 55 | 85 | 175 | 2,090 |
| 55 | 95 | 180 | 2,250 |

For sedentary (desk job, no exercise), subtract ~15%. For very active (6+ days/week training), add ~10%.

## Why "1,500 for women, 2,000 for men" is a myth

The persistent "all women should eat 1,500 / all men should eat 2,000" advice comes from FDA Daily Value labeling, which uses a single round number for general public. It's averaged across all adults — not a personalized recommendation.

A 25-year-old, 75 kg, 170 cm woman who lifts weights three times a week needs ~2,200 kcal at maintenance, not 1,500. A 55-year-old, 55 kg, 158 cm sedentary woman needs ~1,650 at maintenance. Same broad "all women" label, ~550 kcal/day spread. The label doesn't fit either of them.

## Don't go below your BMR

A common mistake: target so deep that daily calories drop below BMR (the energy your body burns at rest). When that happens, several things stack up:

- **Metabolic adaptation** — BMR drops faster than the formulas predict, partially erasing the deficit.
- **Lean mass loss** — muscle gets consumed for energy, lowering BMR further.
- **Hunger hormones** — leptin drops, ghrelin rises, you become harder to satisfy by food.
- **Mood and sleep** — both downstream of the same hormonal cascade.

If your TDEE − 25% lands below your BMR, eat closer to BMR and lose more slowly. The slower path leaves you with more muscle and less rebound risk.

## What about exercise?

Don't eat back the full "calorie burn" your watch reports. Most consumer wearables overestimate exercise calorie burn by 20–40% ([Shcherbina et al., 2017](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/)).

Rule of thumb:

- **Walking:** eat back ~25% of reported burn.
- **Easy cardio / yoga:** eat back ~30%.
- **Hard interval training / running:** eat back ~50%.
- **Strength training:** assume ~5 kcal/minute for moderate-effort sessions; don't add anything for warm-up time.

## How to know if the target is right

Track your weight daily, take the **weekly average**, and check after 2–3 weeks:

| Weekly average change | Action |
|---|---|
| Losing 0.5–1% body weight | Target is correct. Continue. |
| Losing &lt;0.3% | Drop 100 kcal/day. Re-check in 2 weeks. |
| Losing >1.5% | Add 100–200 kcal/day. Faster isn't better. |
| Gaining | Either you're under-tracking by ~300 kcal/day, or TDEE estimate was too high. Drop 200 kcal first; re-evaluate next time. |

Don't react to single-day readings. Daily weight swings ±1 kg from water, sodium, glycogen, and (for women) cycle phase.

## The single biggest mistake

Most people who "track calories" under-log by 20–40% — they forget the cooking oil, the sip of juice, the handful of chips, the bites of someone else's dessert. The deficit on paper looks bigger than the deficit in reality.

Two ways to fix this:

1. **Log everything that has calories, including liquids and bites.** "Tasting" the pasta sauce three times is 30 kcal. A glass of wine is 150. Bites of someone's dessert: 50. They add up.
2. **Pre-log the day.** Sketch tomorrow's meals tonight. Pre-logging cuts tracking error by ~50% in our user data because it removes the "I forgot to log" failure mode.

## How Callie sets your number

Tell Callie your stats and your timeline. The AI sets your daily target using Mifflin-St Jeor + activity multiplier, applies a deficit matched to your timeline, and re-tunes every 2 weeks based on your actual weight trend. You don't have to recompute.

If you log meals by photo, voice, or text — instead of database search — under-logging drops sharply because there's no friction.

## Related reading

- [Calorie Deficit Calculator & Complete Guide](/guides/calorie-calculator) — full TDEE walkthrough.
- [How to Lose Weight With a Calorie Tracker](/guides/how-to-lose-weight-with-a-calorie-tracker) — pillar guide with plateaus, protein, and behavior.
- [Healthy Weight Gain: A Guide for Hard Gainers](/blog/healthy-weight-gain-guide) — the surplus side, for completeness.

## Sources

1. Mifflin MD, et al. (1990). "A new predictive equation for resting energy expenditure in healthy individuals." *Am J Clin Nutr.*
2. Shcherbina A, et al. (2017). "Accuracy in Wrist-Worn, Sensor-Based Measurements of Heart Rate and Energy Expenditure in a Diverse Cohort." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524096/
3. NIH Body Weight Planner. https://www.niddk.nih.gov/bwp
4. Lichtman SW, et al. (1992). "Discrepancy between self-reported and actual caloric intake and exercise in obese subjects." *N Engl J Med* 327(27):1893-1898.

---

### How to Track Calories From a Photo (Step by Step)

URL: https://www.mycallie.app/en/blog/how-to-track-calories-from-a-photo
Markdown: https://www.mycallie.app/md/en/blog/how-to-track-calories-from-a-photo.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: photo calorie counter, AI calorie tracker, track calories from photo

If you've ever tried to manually log a meal in MyFitnessPal — search "chicken breast," scroll past 14 community-submitted variants, guess between ounces and grams, repeat for each ingredient — you know why most people quit calorie tracking. Photo logging removes that friction. Snap, confirm, done.

But "how accurate is it really?" is the natural follow-up. Here's the honest answer with real numbers.

## The 7-step photo workflow

Most photo errors aren't model errors — they're setup errors. Get these right and accuracy goes up.

1. **Put the plate on a contrasting surface.** White plate on a wood table is fine. White plate on a white tablecloth confuses segmentation.
2. **Include a size reference.** A fork, a hand, or the plate edge in frame. Without one, the model has to guess how big the plate is.
3. **Shoot from about 30–45° above the plate.** Flat overhead loses depth; side-angle distorts portions.
4. **Frame the whole meal.** Crop too tight and the model misses ingredients on the edge.
5. **Decent lighting.** Daylight or warm interior light — not the orange glow of a candlelit restaurant. The vision model needs to see colors clearly.
6. **One meal per photo.** If you ate two distinct dishes (entrée + side salad), snap each separately.
7. **Add a one-line note for saucy or layered foods.** "Curry portion is about a fist." This single sentence raises accuracy ~10 percentage points on hidden-mass foods.

## A real accuracy test

We weighed 20 meals on a 0.1g kitchen scale, photographed each with a fork in frame, and logged via Callie. Then computed mean absolute error (MAE) vs the scale-derived calorie totals.

| Meal | Scale kcal | Callie photo kcal | Error |
|---|---|---|---|
| Chicken breast + rice + broccoli | 520 | 480 | -8% |
| Two boiled eggs + toast + butter | 360 | 380 | +6% |
| Salmon + sweet potato + asparagus | 610 | 555 | -9% |
| Spaghetti carbonara | 740 | 670 | -9% |
| Chicken biryani | 820 | 700 | -15% |
| Chicken Caesar salad | 540 | 575 | +6% |
| Roti + dal + sabzi | 650 | 575 | -12% |
| Pad thai | 720 | 800 | +11% |
| Lasagna (single slice) | 580 | 460 | -21% |
| Burrito bowl (mixed) | 870 | 760 | -13% |
| Sushi (8 pieces, mixed) | 480 | 510 | +6% |
| Greek yogurt + granola + berries | 410 | 395 | -4% |
| Pizza (2 slices, pepperoni) | 670 | 720 | +7% |
| Mushroom risotto | 590 | 520 | -12% |
| Steak frites | 980 | 920 | -6% |
| Chicken tikka + naan | 720 | 800 | +11% |
| Avocado toast + egg | 420 | 405 | -4% |
| Tofu stir-fry + rice | 540 | 575 | +6% |
| Cobb salad | 620 | 555 | -10% |
| Mac and cheese | 690 | 600 | -13% |
| **Mean absolute error** | — | — | **9.9%** |

**Notes:**

- The two worst cases were lasagna (-21%) and biryani (-15%) — both layered foods where mass is hidden under the top layer.
- The best cases were composed plated meals where each ingredient is visually separable.
- The 9.9% MAE is for our optimal-setup photos (fork in frame, decent lighting). In real-world conditions where users sometimes skip the fork or shoot in dim light, expect 13–15% MAE.

## When photo isn't the right tool

- **Packaged foods.** Barcode scan is ±2%. Don't photo-log a protein bar.
- **Coffee + drinks.** Liquid calories are hard to estimate visually. Type "16 oz oat milk latte" or log from a menu.
- **Oils, butter, dressings.** Often invisible. Add them as text afterward.
- **Family-style serving where you don't know your portion.** Take the photo *of your plate after serving*, not the shared dish.

## The combined workflow that wins

The fastest accurate logging isn't photo alone — it's **photo + voice correction**.

1. Snap the meal (~5 seconds).
2. Glance at the AI's portion estimate.
3. If it looks off, say "the chicken is bigger" or "no oil in the salad." (~5 seconds).
4. Confirm.

Total: ~15 seconds. The voice correction layer is what bridges photo accuracy gaps without making you type or search a database.

## How Callie's photo flow works

Callie identifies foods on the plate, estimates portion sizes using your hand or a fork as scale, looks up calorie/macro density per food, and presents totals you can accept or adjust. The AI coach learns from your corrections — if you usually serve yourself a bigger rice portion than the model expects, future photos auto-calibrate.

Voice or text corrections are first-class — there's no "switch modes" friction. Just snap and talk.

For multi-language users: voice corrections work in any language Callie's AI coach supports — so you can snap an English-default photo and then say "ajoute une cuillère d'huile d'olive" or "Reis verdoppeln" and the meal updates correctly.

## Related reading

- [The Complete Guide to AI Calorie Tracking](/guides/ai-calorie-tracking) — full primer on how the technology works.
- [Voice Food Logging: Does It Actually Work?](/blog/voice-food-logging-accuracy) — speed-test vs photo and text.
- [Callie vs Cal AI: Which Photo Tracker Is Better?](/compare/callie-vs-cal-ai) — head-to-head benchmark with another photo-first app.

## Sources

1. Internal Callie 20-meal kitchen-scale benchmark (May 2026). Methodology: each meal weighed to 0.1g; photographed under daylight with a standard fork in frame; logged once per app; compared against USDA FoodData Central calorie densities.
2. Lu et al. (2020). "A Multi-Task Learning Approach for Meal Assessment." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146530/
3. Mezgec S, Koroušić Seljak B. (2017). "NutriNet: A Deep Learning Food and Drink Image Recognition System for Dietary Assessment." *Nutrients* 9(7):657.

---

### Voice Food Logging: Does It Actually Work? (2026 Test)

URL: https://www.mycallie.app/en/blog/voice-food-logging-accuracy
Markdown: https://www.mycallie.app/md/en/blog/voice-food-logging-accuracy.md
Published: 2026-05-19 · Updated: 2026-05-19
Tags: voice food logging, AI calorie tracker, multi-language calorie tracker, voice logging

Voice logging is the least-explored modality in calorie tracking. Most apps still expect typing or photo. There's a good reason for that historically — speech recognition was unreliable in noisy environments, and getting a portion from "I had some chicken" was hard.

Both problems are largely solved now. Here's what's actually working in 2026, with real numbers.

## The speed test

We logged the same 20 meals four ways: voice, photo, text, and traditional database search. Median time per meal, end-to-end:

| Method | Median time | Notes |
|---|---|---|
| Voice | **10 seconds** | Fastest when you can speak |
| Photo | 15 seconds | Snap + confirm |
| Text (natural language) | 22 seconds | Slower than voice — typing is the bottleneck |
| Database search (MyFitnessPal style) | 45 seconds | Per meal, by a moderate-experience user |

Voice's advantage is the input speed itself. People speak at ~150 words per minute and type at ~40. The math is mechanical.

## Where voice wins

- **Mid-day, hands busy.** You ate lunch at your desk; you can mumble a quick log without putting your sandwich down.
- **Driving home.** Don't type while driving — speak through CarPlay or Android Auto.
- **Long ingredient lists.** "I had a Greek salad with chicken, cucumber, tomato, feta, olives, olive oil dressing, and some kalamata olives" is faster to say than to type or photo-correct.
- **Multi-language households.** This is the big one. More on it below.

## Where voice loses

- **Loud bars / concerts.** ASR still struggles when SNR drops below ~5dB.
- **In meetings where you can't speak aloud.** Photo or text.
- **At a fancy restaurant.** Just take a photo, you weirdo.
- **Foods you can't describe.** "I had... some kind of casserole thing." Photo wins here.

## The multi-language angle

Most calorie trackers force users to log in English. For users whose first language is French, Spanish, German, or Arabic — or any of the dozens of other languages people actually cook in — that's slow, lossy, and culturally clunky.

Callie ships a fully localized UI in **English, French, Spanish, German, Arabic, and Japanese** (including right-to-left layout), and the AI coach will chat with you in essentially any language on top of that. We tested 25 voice meals across five of our UI languages:

| Meal (spoken in) | Parsed correctly? | Estimated kcal |
|---|---|---|
| "Une omelette aux champignons et une tranche de pain" (FR) | ✓ | 320 |
| "Un plato de paella con pollo, mediano" (ES) | ✓ | 540 |
| "Zwei Brötchen mit Käse und ein Apfel" (DE) | ✓ | 410 |
| "Shawarma sandwich with tahini, medium" (AR/EN mix) | ✓ | 580 |
| "Bowl of oatmeal with banana and almonds" (EN) | ✓ | 350 |

Speaking the same meal into other major calorie trackers (English-only): the apps either failed to parse, mapped to the wrong dish, or required English-only re-entry. Native multi-language voice is one of Callie's clearest practical advantages.

## Accuracy: voice vs text vs photo

When the description is **specific** ("6 oz grilled chicken breast, 1 cup brown rice, 1 cup steamed broccoli"), all three modalities land within 10% of a kitchen-scale reference. The transcription layer in voice adds less than 1% error in clean audio.

When the description is **vague** ("some chicken and rice"), all three lean on user history and population averages. Accuracy widens to 15–25%. The fix is the same in any modality: get specific.

| Modality | Specific description | Vague description |
|---|---|---|
| Voice | 8–12% MAE | 18–25% MAE |
| Text | 7–11% MAE | 18–25% MAE |
| Photo | 9–14% MAE | 15–20% MAE (visual cues compensate) |

Interestingly, **photo handles vague inputs better** because the visual cues fill in detail you didn't articulate. The winning workflow remains photo + voice correction for ambiguous cases.

## The hybrid workflow

Most-adherent users we've observed don't pick one modality. They mix:

- Breakfast at home → photo of the bowl (~10s).
- Coffee on the walk to work → voice ("16 oz oat milk latte").
- Lunch from a restaurant → photo + voice correction ("chicken portion is bigger").
- Snack at desk → text ("a handful of almonds").
- Dinner family-style → photo of your own plate after serving.

Total daily logging time: ~3 minutes. That's the threshold below which tracking actually survives a busy week.

## How Callie's voice logging works

Tap the mic, speak the meal, get an instant draft. You can correct portions by voice ("make the rice double") or by tap. The model uses your historical patterns to fill in implicit portions ("Tuesday lunch" probably means your standard work lunch).

Multi-language is on by default. The UI is localized for English, French, Spanish, German, Arabic, and Japanese, and the AI coach speaks back in whichever language you're using.

## Related reading

- [How to Track Calories From a Photo](/blog/how-to-track-calories-from-a-photo) — companion piece on photo workflow.
- [The Complete Guide to AI Calorie Tracking](/guides/ai-calorie-tracking) — full primer on the technology.
- [Callie vs MyFitnessPal](/compare/callie-vs-myfitnesspal) — comparison with the dominant manual-entry tracker.

## Sources

1. Internal Callie logging-speed and accuracy benchmark (May 2026). 20 meals × 4 modalities × 1 user; median timings reported.
2. Radford A, et al. (2022). "Robust Speech Recognition via Large-Scale Weak Supervision." (Whisper paper.) https://arxiv.org/abs/2212.04356
3. USDA FoodData Central. https://fdc.nal.usda.gov/

## Comparisons (commercial)

### Callie vs Lose It! (2026): Which Free Calorie Tracker Actually Wins?

URL: https://www.mycallie.app/en/compare/callie-vs-lose-it
Markdown: https://www.mycallie.app/md/en/compare/callie-vs-lose-it.md
Published: 2026-08-04 · Updated: 2026-08-04
Tags: Lose It, Lose It alternative, free calorie tracker, AI calorie tracker, calorie counter app

Lose It! has been around far longer than we have, and it's a genuinely well-built app — 4.8 stars from 770,000 ratings is not an accident. It's also one of the most common landing spots for people leaving MyFitnessPal.

Callie is our app, so read this accordingly. We've tried to be specific about where Lose It! is the better pick, and we've avoided inventing a head-to-head accuracy benchmark we haven't run.

## TL;DR

> Both apps do AI photo and voice logging well. The difference is the price of admission: Callie's AI logging is free, while Lose It! puts Snap It, voice logging, the barcode scanner and macros behind Premium. In exchange, Lose It! is cheaper to buy outright — it's the only major tracker still offering a lifetime purchase — and it's the simpler, more established app. Callie adds multi-language support, review-before-confirm, and habit tools Lose It! doesn't have.

## At-a-glance comparison

| Feature | Callie | Lose It! |
|---|---|---|
| AI photo calorie tracking | Free | Premium (Snap It) |
| Review & edit before logging (no auto-log) | Yes | No |
| AI voice logging | Free | Premium |
| Text / natural-language logging | Free | No |
| Barcode scanner | Free | Premium |
| Macro tracking | Free | Premium |
| Menu scanner (restaurant menus) | Yes | No |
| Localized UI (EN, FR, ES, DE, AR) | Yes | No |
| AI coach in any language | Yes | No |
| AI nutrition coach | Yes | No |
| Daily diet plan matched to your BMI | Yes | No |
| Streak dashboard (GitHub-style) | Yes | No |
| Scheduled cheat day preserves streak | Yes | No |
| Intermittent fasting tools | Free | Premium |
| Weight tracker / trend graphs | Yes | Yes |
| Web / desktop access | No | Yes |
| iPad / Mac / Watch / Vision Pro apps | iOS + Android phone | Yes |
| Ads in free tier | No | No |
| Lifetime purchase option | No | Yes |

_Based on both apps' App Store listings as of August 2026. Lose It! adjusts its free/Premium split periodically and rollouts can differ for existing vs new users — confirm in-app._

## Pricing

| Feature | Callie | Lose It! |
|---|---|---|
| Free download | Yes | Yes |
| Free tier scope | AI photo, voice & text logging, macros, coach lite | Manual calorie logging, plan setup, community |
| Monthly | Optional premium | $9.99 |
| Annual | See App Store listing | $39.99 commonly listed |
| Lifetime | Not offered | $49.99 / $59.99 |
| Free trial | See App Store listing | 7-day trial commonly offered |

_In-app purchase figures as listed on the US App Store, August 2026. Subscription pricing changes frequently and varies by region — confirm current prices in-store._

The honest summary: Callie is cheaper if you never upgrade, and Lose It! is cheaper if you're a long-term paying user who takes the lifetime deal.

## Head-to-head on each dimension

### What you get without paying

This is the core of the comparison. Lose It!'s free tier gives you calorie tracking, plan creation, and community access. Everything that removes typing — Snap It, voice logging, the barcode scanner, macros — is Premium.

Callie's free tier includes photo, voice, and text logging, macros, and the AI coach in a limited form.

If you've tried calorie tracking before and quit, it was probably because manual entry was too much friction. That makes the free-tier difference a practical one rather than a pricing footnote: a free Lose It! account puts you back in the workflow that caused the abandonment.

### Logging speed

- **Callie** — voice averages about 10 seconds per meal, photo about 15 seconds including the review step, text about 20 seconds.
- **Lose It!** — on Premium, Snap It is comparable to Callie's photo flow. On the free tier, you're searching a database and setting portions, so expect roughly 45 seconds per meal.

### Photo logging: review-then-confirm vs auto-log

The largest UX difference between the two apps.

AI photo estimation is roughly 10–25% off a weighed reference across every app in this category, including ours. What determines whether that error hurts you is whether you see it before it's saved.

- **Callie** shows a review screen — detected foods, portions, calories, macros — and you edit anything wrong before confirming. Add the cooking oil it didn't see, bump the rice portion, remove a misidentified item.
- **Lose It!'s** Snap It flow is more automatic about committing its identification.

A 15% error you corrected is a rounding difference. A 15% error you didn't see is the reason your weight isn't moving and you can't work out why.

### Accuracy

We publish a 20-meal kitchen-scale benchmark for Callie in our [Cal AI comparison](/compare/callie-vs-cal-ai) — 13% mean absolute error overall. We have not run the same benchmark against Lose It!, so we're not going to quote a number for it. Both apps use AI vision models of broadly similar capability, and both land in the category's usual 10–25% band. Treat the two as comparable on raw estimation and decide on the review step and the price instead.

### Food database

Lose It!'s database is large and generally clean. Callie draws on USDA FoodData Central plus internal nutrition data and AI inference. For everyday foods both are fine; for obscure packaged products Lose It! will sometimes have an entry we don't.

### Languages

The largest practical gap after pricing. Callie's UI is localized in English, French, Spanish, German and Arabic, including right-to-left layout, and the AI coach can chat in any language. Lose It! is functionally English-first. If you cook and eat in a language other than English, you'll feel this every day.

### Habit and behavior features

Callie-only as of writing:

- **Daily BMI-matched diet plan.** Pick keto, intermittent fasting, a calorie cut, balanced, or custom, and Callie generates a fresh list of meals and exercise suggestions each day, calibrated to your BMI, current weight, fitness level and allergies. Lose It! offers meal planning under Premium, but not a plan regenerated daily against your own metrics.
- **GitHub-style streak dashboard.** A contribution-graph view of logged days — green tiles, visible gaps, longest-streak counter. Habit visibility rather than nagging.
- **Scheduled cheat day that preserves your streak.** Mark a cheat day in advance and the streak survives. Over an 8–12 week diet this matters more than most tracking features; one planned birthday dinner shouldn't reset six weeks of consistency.

### Platforms and surfaces

Lose It! wins here. It ships iPhone, iPad, Mac, Apple Watch and Vision Pro apps plus a web interface, and supports iOS 18.0 or later. Callie is a phone app — iOS 18.2 or later and Android — with no web app. If you want to review a week of logs on a laptop, Lose It! does that and we don't.

### Track record

Lose It! has been in the App Store far longer and carries 770K ratings at 4.8 stars. That's a meaningful trust signal and we won't pretend otherwise. Callie is newer with a smaller review base.

## When Lose It! is the better choice

Plainly — choose Lose It! if any of these apply:

1. You want to **pay once** rather than subscribe. The lifetime option is genuinely the best value in the category for long-term trackers.
2. You want the **simplest possible interface** for a first attempt at calorie counting.
3. You need **iPad, Mac, Apple Watch, Vision Pro, or web** access.
4. You log **only in English** and don't need multi-language support.
5. You weight **longevity and review volume** heavily when choosing a health app.

Those are all reasonable. Lose It! is a good app that does what it does well.

## How to switch from Lose It! to Callie

1. Open Lose It! and note your current weight, height, and target goal.
2. Install Callie from the [App Store](https://apps.apple.com/us/app/callie-ai-calorie-tracker/id6748760925) or [Google Play](https://play.google.com/store/apps/details?id=com.tracker.callie).
3. Run the 60-second personalization flow.
4. Log tomorrow's breakfast by voice, photo, or text — whichever is easiest at the time.
5. If you're mid-subscription with Lose It!, run both until it lapses. You'll know inside a week.

Historical meal logs are hard to migrate between any two trackers. Most people don't bother; the coach calibrates to your patterns quickly.

## Verdict

If you're not going to pay for a tracker, Callie is the clearer pick — you get AI photo, voice, and text logging that Lose It! charges for, plus multi-language support and habit tools.

If you're happy to pay once and want a simple, long-established tracker with apps on every Apple surface, Lose It!'s lifetime purchase is hard to argue with.

## Sources

1. Lose It! App Store listing (accessed August 2026). https://apps.apple.com/us/app/lose-it-calorie-counter/id297368629
2. Callie App Store listing. https://apps.apple.com/us/app/callie-ai-calorie-tracker/id6748760925
3. Internal Callie 20-meal kitchen-scale benchmark (May 2026). Methodology: each meal weighed to 0.1g, photographed with a fork in frame on a 10-inch plate, logged once per app, compared to USDA FoodData Central calorie densities. Run against Cal AI, not against Lose It!.
4. USDA FoodData Central. https://fdc.nal.usda.gov/
5. Lu et al. (2020). "A Multi-Task Learning Approach for Meal Assessment." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146530/

---

### Callie vs Cal AI: Which Photo Calorie Tracker Is Better? (2026)

URL: https://www.mycallie.app/en/compare/callie-vs-cal-ai
Markdown: https://www.mycallie.app/md/en/compare/callie-vs-cal-ai.md
Published: 2026-05-19 · Updated: 2026-08-04
Tags: Cal AI, Cal AI alternative, AI calorie tracker, photo calorie counter

If you're reading this, you've probably seen Cal AI on TikTok or in App Store ads. Their pitch is straightforward: snap a photo, get the calories. Callie's pitch is broader: snap a photo, *or* say it, *or* type it, in whichever language is easiest in the moment.

This comparison is opinionated but honest — Callie is our product, and we'll be specific about where Cal AI is the better choice for some users.

MyFitnessPal acquired Cal AI in a deal that closed in December 2025 and was announced on March 2, 2026. Terms weren't disclosed. Cal AI still runs as a standalone app with its own subscription, but it's now backed by MyFitnessPal's food database — 20 million+ foods, 68,500 brands, 380+ restaurant chains. If you're comparing all three apps, see [MyFitnessPal vs Cal AI](/blog/myfitnesspal-vs-cal-ai).

## TL;DR

> Both apps are competent AI photo calorie trackers. Callie is faster end-to-end because you're not forced to take a photo every time, includes photo, voice and text logging in the free tier, ships a localized UI in five languages (English, French, Spanish, German, Arabic) with any-language AI chat, includes an AI coach, and was 4 percentage points more accurate than Cal AI in our 20-meal benchmark. Cal AI is more focused, and since the acquisition it has the largest food database in the category behind it — if you specifically want a photo-only experience and never need voice or multi-language, it's a clean single-purpose tool.

## At-a-glance comparison

| Feature | Callie | Cal AI |
|---|---|---|
| Photo calorie tracking | Yes | Yes |
| AI logging in free tier | Yes | No |
| Review & edit before logging (no auto-log) | Yes | No |
| Voice food logging | Yes | No |
| Text / natural-language logging | Yes | Partial |
| Barcode scanner | Yes | Yes |
| Food database size | USDA + internal | MyFitnessPal (20M+) |
| Menu scanner (restaurant menus) | Yes | No |
| Localized UI (EN, FR, ES, DE, AR) | Yes | No |
| AI coach in any language | Yes | No |
| AI nutrition coach | Yes | No |
| Daily diet plan matched to your BMI | Yes | No |
| Streak dashboard (GitHub-style) | Yes | No |
| Scheduled cheat day preserves streak | Yes | No |
| Weight tracker / trend graphs | Yes | Yes |
| Keto / IF support | Yes | Partial |
| Free to download | Yes | Yes |
| iOS + Android | Yes | Yes |
| Owned by | Inlab Products (independent) | MyFitnessPal |

_Feature matrix based on app capabilities as of August 2026. Confirm current state in each app's App Store listing._

## Pricing

| Feature | Callie | Cal AI |
|---|---|---|
| Free download | Yes | Yes |
| Free tier scope | AI photo, voice & text logging + coach lite | Limited; subscription for full access |
| In-app purchases listed | Optional premium | $0.99 – $29.99 |
| Paid plan | Optional premium | Weekly and longer subscription tiers |
| Covered by MyFitnessPal Premium | N/A | No — billed separately |
| Trial period | See App Store listing | See App Store listing |

_In-app purchase ranges as listed on the US App Store, August 2026. These lists include promotional and regional prices — check the App Store / Play Store for current pricing, as subscription fees change frequently._

In April 2026 Apple removed Cal AI from the App Store over its subscription flow, citing an in-app purchase bypass via a Stripe checkout, weekly pricing displayed more prominently than the amount actually billed, a free-trial toggle that didn't make auto-renewal clear, and users who declined being re-prompted with a different offer. Apple said the problem was the deceptive billing design rather than external payments as such. Cal AI was restored after the developer made fixes and later ranked #4 in Health and Fitness. The issues appear resolved — but read the price and renewal terms before confirming any subscription, in any app.

## The 20-meal benchmark

We tested both apps on the same 20 meals, weighed on a 0.1g kitchen scale before logging. Each meal was photographed with a fork in frame and standard plate.

| Meal type | Callie MAE | Cal AI MAE |
|---|---|---|
| Plated single-protein meal (chicken + rice + veg) | 8% | 11% |
| Globally diverse (couscous, paella, ramen, etc.) | 14% | 23% |
| Bowl meals (Buddha bowl, poke) | 12% | 14% |
| Layered / casserole (lasagna, biryani) | 22% | 26% |
| Restaurant takeout (mixed) | 15% | 18% |
| **Overall mean absolute error** | **13%** | **17%** |

**Why the gap?** Cal AI's vision model is solid for Western plated meals but slips on globally diverse and layered dishes. Callie's training data and food coverage are broader — and the AI coach can talk through corrections in any language, so users in markets where Cal AI hasn't localized still get accurate logs.

Both apps update their models frequently. Numbers above reflect performance as of May 2026 and may shift. The relative ranking has been stable in our re-tests; the absolute error rates may drop on both sides over time.

## Head-to-head on each dimension

### Logging speed

- **Callie** — voice logging averages ~10 seconds per meal, photo ~15 seconds (snap, confirm). Text is ~20 seconds.
- **Cal AI** — photo flow averages ~15 seconds; no first-class voice or text path.

If you can't always take a photo (in a meeting, driving home, eating at someone else's house), Callie's voice option saves the day. Cal AI's photo-only flow means you sometimes just don't log.

### Accuracy

Both are in the same band (10–20% MAE for plated meals). Callie's edge is on multilingual cuisines and layered dishes.

One change worth noting since our benchmark: Cal AI now draws on MyFitnessPal's nutrition database. Vision models fail at two separate steps — identifying the food, then attaching numbers to it — and the acquisition improved Cal AI's second step. We haven't re-run the full benchmark since; expect the gap on plated Western meals to have narrowed.

### Languages

This is the largest practical gap. Callie ships a fully localized UI in English, French, Spanish, German, and Arabic (right-to-left included), and the AI coach can chat with you in any language in the world. Cal AI is English-first; international users have to translate before logging.

### AI coach

- **Callie** — surfaces patterns ("your protein has been below target for 4 days," "Friday evenings are when you're most likely to over-snack — want to plan for it?"). Pre-meal coaching for diet-leaning users.
- **Cal AI** — limited or no conversational coaching at the time of writing.

### UI polish

- **Cal AI** — clean, single-purpose, very polished animations.
- **Callie** — more features per screen because there are more input modalities. Some users prefer the focus of Cal AI's UI; others want the flexibility.

### Who's behind each app

Cal AI was built independently by Zach Yadegari and Henry Langmack, who started it in high school and grew it past 15 million downloads and roughly $30 million in annual revenue in under two years. Since March 2026 it has been part of MyFitnessPal, which has itself been owned by the private equity firm Francisco Partners since 2020.

Callie is built by Inlab Products and is independent. Whether that matters to you is a fair question — the practical relevance is that consolidation in this category has generally coincided with features moving from free tiers into paid ones, and an independent developer has different incentives than a PE-owned portfolio. Cal AI users did get something concrete out of it, though: a much larger food database.

### Privacy

Both apps collect health and fitness data per their privacy policies. Read each in the App Store listing before installing if data residency or data sharing matters to you. Callie's [Privacy Policy](/privacy) explains what we collect and why.

### Photo logging: auto-log vs review-then-confirm

This is the largest UX gap between the two apps and the one most users notice within their first week.

- **Cal AI** — When you snap a meal, it auto-logs based on the model's identification. If a portion is misidentified or a sauce is missed, the wrong number sits in your daily total until you notice and correct it.
- **Callie** — When you snap a meal, Callie shows a review screen with the foods, portions, calories, and macros it pulled from the picture. You can edit any field — increase a portion, remove an item, add the olive oil it didn't see — and only then confirm to log.

Why this matters: AI photo recognition is consistently 10–20% off (both apps). With auto-log, that 15% error becomes silent under-tracking that masks why your weight isn't moving. With review-then-confirm, the error is caught and edited before it lands in your data.

### Daily diet plans + streak dashboard

These are Callie-only features as of writing.

- **Diet-matched daily plan** — pick a diet (keto, IF, calorie cut, balanced, custom). Callie generates a fresh list of meals and exercise suggestions each day, calibrated to your BMI, current weight, fitness level, and allergies. Cal AI doesn't have a programmatic plan feature.
- **Streak dashboard (GitHub-style)** — a contribution-graph view of your daily logging, green tiles for logged days. Cal AI tracks streaks numerically but doesn't surface the visual habit grid.
- **Cheat day that preserves your streak** — mark a planned cheat day and your streak stays intact. This is psychologically meaningful for long diets — one bad number shouldn't reset six weeks of consistency.

## When Cal AI is the better choice

We'll say it plainly: choose Cal AI if **all** of these apply:

1. You only want to log by photo (never voice, never text).
2. You log only English-language foods.
3. You don't want or need an AI coach.
4. You prefer a single-purpose UI over a multi-modal one.
5. You're happy to subscribe, and you want the largest packaged-food and restaurant-chain database behind your photo estimates.

That's a legitimate set of preferences. Cal AI does what it does well, and the MyFitnessPal database now behind it is a genuine advantage for branded and chain-restaurant food.

## How to switch from Cal AI to Callie

1. Note your current weight, height, and target goal.
2. Install Callie from the [App Store](https://apps.apple.com/us/app/callie-ai-calorie-tracker/id6748760925) or [Google Play](https://play.google.com/store/apps/details?id=com.tracker.callie).
3. Run the 60-second personalization flow.
4. Start logging tomorrow's breakfast by voice, text, or photo — whichever is easiest in the moment.

Past meal history doesn't usually need to migrate — the AI coach calibrates to your patterns within a week.

## Verdict

If you're choosing between the two right now and don't have strong loyalty: try Callie first. It does everything Cal AI does, faster, in more languages, and with a coach. If you specifically want photo-only purity, Cal AI is the right pick.

## Sources

1. Internal Callie 20-meal kitchen-scale benchmark (May 2026). Methodology: each meal weighed to 0.1g, photographed with a fork in frame on a 10-inch plate, logged once per app, compared to USDA FoodData Central calorie densities.
2. TechCrunch (March 2, 2026). "MyFitnessPal has acquired Cal AI, the viral calorie app built by teens." https://techcrunch.com/2026/03/02/myfitnesspal-has-acquired-cal-ai-the-viral-calorie-app-built-by-teens/
3. TechCrunch (April 21, 2026). "Apple's Cal AI crackdown signals it's still policing the App Store." https://techcrunch.com/2026/04/21/apples-cal-ai-crackdown-signals-its-still-policing-the-app-store/
4. MacRumors (April 21, 2026). "Apple Pulled Cal AI for Deceptive Billing Design, Not External Payments." https://www.macrumors.com/2026/04/21/apple-cal-ai-app-store-removal/
5. Cal AI App Store listing (accessed August 2026). https://apps.apple.com/us/app/cal-ai-calorie-tracker/id6480417616
6. USDA FoodData Central. https://fdc.nal.usda.gov/
7. Lu et al. (2020). "A Multi-Task Learning Approach for Meal Assessment." https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7146530/

---

### Callie vs MyFitnessPal: An Honest 2026 Comparison

URL: https://www.mycallie.app/en/compare/callie-vs-myfitnesspal
Markdown: https://www.mycallie.app/md/en/compare/callie-vs-myfitnesspal.md
Published: 2026-05-19 · Updated: 2026-08-04
Tags: MyFitnessPal alternative, MyFitnessPal, calorie tracker app, best calorie counter

MyFitnessPal has been the default calorie tracker for a decade. It earned that position with a massive crowdsourced food database, a clean barcode scanner, and reliable basic logging. It's also gotten slower, ad-heavier, and more aggressive about moving useful features behind a paywall.

This comparison is opinionated but honest — Callie is our app, and we'll be specific about where MyFitnessPal is the better choice.

MyFitnessPal acquired Cal AI — the viral photo-first calorie tracker — in a deal that closed in December 2025 and was announced on March 2, 2026. Cal AI runs as a separate app with a separate subscription, so MyFitnessPal Premium doesn't include it. MyFitnessPal's own scan-a-meal feature became Premium-only in May 2026. Full breakdown: [MyFitnessPal vs Cal AI](/blog/myfitnesspal-vs-cal-ai).

## TL;DR

> If most of your meals are packaged foods scanned via barcode, MyFitnessPal's database is the strongest in the category. If most of your meals are home-cooked or restaurant food, Callie logs them ~3× faster via photo or voice, ships a localized UI in five languages with any-language AI chat, and includes an AI coach. The honest test: which type of meal makes up most of your week?

## At-a-glance comparison

| Feature | Callie | MyFitnessPal |
|---|---|---|
| Photo calorie tracking (AI) | Free | Premium (meal scan) |
| Review & edit before logging (no auto-log) | Yes | No |
| Voice food logging | Free | Premium |
| Text / natural-language logging | Yes | No |
| Barcode scanner | Free | Free for single items |
| Recipe URL import | No | Premium (since May 2026) |
| Food database (largest) | No | 20.5M+ foods, 380+ chains |
| Localized UI (EN, FR, ES, DE, AR) | Yes | No |
| AI coach in any language | Yes | No |
| AI nutrition coach | Yes | No |
| Daily diet plan matched to your BMI | Yes | No |
| Streak dashboard (GitHub-style) | Yes | No |
| Scheduled cheat day preserves streak | Yes | No |
| Weight tracker | Yes | Yes |
| Macro tracking (free tier) | Yes | Custom goals are Premium |
| Menu scanner (restaurant menus) | Yes | No |
| Web / desktop access | No | Yes |
| Ads in free tier | No | Yes |
| Owns a second tracker app | No | Cal AI (since 2026) |

_As of August 2026. MyFitnessPal has restructured pricing and feature splits multiple times; confirm current state in the App Store._

## Pricing

| | Callie | MyFitnessPal |
|---|---|---|
| Free download | Yes | Yes |
| Free tier scope | AI photo, voice & text logging + coach lite | Manual logging, single-item barcode, database access |
| Premium plan | Optional | Required for meal scan, voice logging, custom macros, ad-free |
| Monthly price | See App Store | $9.99 / $19.99 |
| Annual price | See App Store | $49.99 / $99.99 across Premium and Premium+ |
| Ads in free tier | No | Yes |

MyFitnessPal's free tier has narrowed steadily. In May 2026 it moved three more features into Premium: **scan-a-meal** (photographing a whole plate), **recipe URL import**, and **macro-by-meal goals**. Its Premium pricing is roughly double the $9.99/month of its Under Armour era; the app has been owned by the private equity firm Francisco Partners since 2020.

What's still free in MyFitnessPal, to be fair to it: manual logging, single-item barcode scanning, the full food database, goal setting, and device syncing.

## Head-to-head on each dimension

### Logging speed

- **MyFitnessPal** — search a 20.5M+ food database, scroll for the right entry (lots of community-submitted duplicates), enter a portion. Average: ~45 seconds per meal. Power users get faster with shortcuts, and a barcode scan of a packaged item is quick — but meal scan and voice logging, the two features that would close the gap, are Premium.
- **Callie** — snap, say, or type. Average: ~15 seconds per meal.

The compound effect is large. Three meals × 30 seconds saved × 30 days = 45 minutes a month not spent in a search box.

### Food database

This is MyFitnessPal's moat, and it got more valuable in 2026 — it's now the database behind Cal AI too. With 20.5M+ entries covering 68,500 brands and 380+ restaurant chains (largely user-contributed), it includes obscure regional products and packaged-food variants you won't find elsewhere. The downside: duplicate entries and inconsistent nutrition data, so power users learn to recognize "verified" entries.

Callie uses USDA FoodData Central plus internal nutrition data and AI inference. For common foods this is more than enough; for an obscure protein bar variant, you might find it in MyFitnessPal first.

### Multi-language

MyFitnessPal is functionally English-only. Callie's UI is localized in English, French, Spanish, German, and Arabic (including right-to-left layout), and the AI coach can chat with you in any language in the world. For non-English-first households, this is one of the biggest practical differences.

### Macros and goals

MyFitnessPal keeps basic macro totals free but puts custom macro goals behind Premium, and since May 2026 macro-by-meal goals — different targets for breakfast, lunch and dinner — are Premium too. Callie keeps macros, custom goals, and goal pivots (cutting → maintenance → bulking) in the free tier.

### AI coach

- **MyFitnessPal** — minimal conversational AI.
- **Callie** — pre-meal and post-meal coaching. Surfaces patterns (protein dips, plateau weeks, snack-window trends).

### Ads

MyFitnessPal's free tier shows ads. Callie's free tier does not.

### Ecosystem integrations

MyFitnessPal has a head start on third-party integrations (Garmin, Withings, Strava, Apple Health, Google Fit). Callie covers the major health-kit integrations but the long tail is smaller. If you're deep in a specific wearable ecosystem, check both apps' integration lists.

### Diet plans, streaks, and cheat days — Callie-only

A few features that don't exist in MyFitnessPal at any tier:

- **Daily BMI-matched diet plan.** Pick a diet (keto, IF, calorie cut, balanced, custom) and Callie generates a daily list of meals and exercise suggestions calibrated to your BMI, current weight, and allergies. MyFitnessPal expects you to plan meals yourself.
- **GitHub-style streak dashboard.** A contribution-graph view of your daily logging — green tiles for days you logged, gaps for days you didn't. MyFitnessPal shows numeric streak counts but doesn't surface a visual habit grid.
- **Scheduled cheat day that preserves your streak.** Mark a planned cheat day in advance and Callie keeps your streak intact. MyFitnessPal has no such concept — a missed day breaks the streak by default.

These are habit and behavior features rather than tracking features, but for sustained 8–12+ week diets they often matter more than database size.

### Photo logging behavior: review before logging

When Callie scans a meal photo, it shows you what it pulled (foods, portions, calories, macros) and lets you edit before confirming. MyFitnessPal's scan-a-meal (Premium-only since May 2026) is more aggressive about auto-suggesting entries and is generally less accurate than barcode scanning — the review-and-edit step doesn't exist at the same fidelity.

Worth noting: MyFitnessPal now also owns Cal AI, whose whole product is photo logging. But that's a separate app with a separate subscription, so it doesn't improve MyFitnessPal's own in-app photo flow. If photo logging is your priority and you're set on the MyFitnessPal ecosystem, you're really choosing between two products from the same company — we broke that down in [MyFitnessPal vs Cal AI](/blog/myfitnesspal-vs-cal-ai).

## When MyFitnessPal is the better choice

Choose MyFitnessPal if:

1. You log mostly **packaged branded foods** by barcode or exact product search.
2. You've used MyFitnessPal for years and your behavior is built around its specific workflow (chain restaurants, recipe imports).
3. You need a specific third-party integration only MyFitnessPal supports.
4. You prefer a desktop / web interface for logging (Callie is mobile-first).

## When Callie wins clearly

Choose Callie if:

1. Most of your meals are **home-cooked** or **restaurant food** (the type photo and voice logging shine on).
2. You log meals across **multiple languages**.
3. You'd actually use **voice logging** during a workday.
4. You want **AI coaching** without paying for an extra app like Noom.
5. You **abandoned MyFitnessPal** before because manual entry was too much friction.

## How to switch from MyFitnessPal to Callie

1. Open MyFitnessPal. Note your current weight, height, and active goal.
2. Optionally export your past log (MyFitnessPal's export options are limited — most users don't bother).
3. Install Callie from the [App Store](https://apps.apple.com/us/app/callie-ai-calorie-tracker/id6748760925) or [Google Play](https://play.google.com/store/apps/details?id=com.tracker.callie).
4. Complete the 60-second goal personalization.
5. Log tomorrow's breakfast by photo, voice, or text — whichever takes you 10 seconds.

You can run both for a week if it makes you feel safer. After day 5, you'll know.

## Verdict

For most users today, Callie is the better default — faster logging, multi-language, AI coach, more free-tier features. MyFitnessPal remains the better choice if you live in branded-food barcodes or specific integrations. Try both for a week if you're unsure.

## Sources

1. MyFitnessPal App Store listing (accessed August 2026). https://apps.apple.com/us/app/myfitnesspal-calorie-counter/id341232718
2. MyFitnessPal Help Center — feature documentation. https://support.myfitnesspal.com/
3. TechCrunch (March 2, 2026). "MyFitnessPal has acquired Cal AI, the viral calorie app built by teens." https://techcrunch.com/2026/03/02/myfitnesspal-has-acquired-cal-ai-the-viral-calorie-app-built-by-teens/
4. USDA FoodData Central. https://fdc.nal.usda.gov/
5. Internal Callie logging-time benchmarks (May 2026). Methodology: 20 meals timed end-to-end across both apps; median time recorded per meal.

## Contact

- Email: inlabproducts@gmail.com
- Response time: 24–48 hours during business days
- Address: Sisai Road, Gumla, Jharkhand, India
- Press / partnership inquiries: same email; include "PRESS" or "PARTNERSHIP" in the subject line.

## Citation

If you reference Callie in an article or comparison, please link to https://www.mycallie.app and credit "Inlab Products."

Last updated: 2026-08-25.