4.8 ★ App Store · 10,000+ users · Cited from USDA & ADA
For carbs, accuracy beats willpower.
C ounting carbs is the single most tedious task in daily diabetes management. The American Diabetes Association recommends carb counting for nearly every patient — but the tools we hand people for the job are still spreadsheets, lookup tables, and the kind of estimation that breaks down at restaurants. The result is what every endocrinologist sees every Monday morning: a week of meals that weren't logged, an A1C that drifted up, and a patient who already knows they're failing.
CalEye approaches the problem from the other side. Instead of asking you to remember the carb content of a chapati, a slice of rye, or a 130-gram serving of dal, we trained a model on tens of thousands of medically-referenced foods to read it from a single photograph. Carbs in grams. Glycemic load. Both backed by USDA SR-Legacy and the ADA Exchange List. Both visible in under a second. Both available offline.
Below: how the system handles the three things diabetics actually need help with — the post-meal check, the restaurant menu, and the long arc of A1C.
GL, not GI. Here's why.
Glycemic index measures sugar speed. Glycemic load measures sugar volume. For meal planning, only the load matters.
- 01 Identify carb grams The model isolates each dish and resolves its carbohydrate content per gram of actual portion — not the standard 100 g reference serving that bears no resemblance to what's on your plate.
- 02 Multiply by glycemic index Each food carries a glycemic index value drawn from peer-reviewed literature and cross-checked against the ADA Exchange List. The carb figure and the GI travel together through the calculation.
- 03 Divide by 100 → glycemic load The result is a single number you can act on. A GL under 10 is low. Between 10 and 20 is moderate. Over 20 warrants attention. No color codes. No ambiguity.
What it looks like, day to day.
Post-prandial check.
The two-hour blood sugar window after eating is the metric most endocrinologists watch most closely. Post-meal glucose variability is larger than most people expect — and highly individual. Photograph the meal you've just eaten and CalEye logs the glycemic load retroactively — before the window closes, before memory softens the portion size. The picture is the log entry. No text input required, no manual gram estimation, no searching a food database for the closest approximation of what you actually ate.
Restaurant menus.
Menu descriptions are optimistic. A "light" pasta dish may carry 80 grams of carbs once you factor in the actual portion and the sauce. CalEye reads the plate you're served — not the item as described — and produces a glycemic load figure based on what the camera can actually see. That distinction matters every time the kitchen is generous, which is almost always. No guesswork, no table-lookup, no mental arithmetic while the waiter waits.
A1C tracking.
Hemoglobin A1C is a 90-day average — it moves from the accumulation of small daily decisions, not from occasional major ones. CalEye's history view surfaces the recurring patterns: the high-GL breakfast that happens every Tuesday, the restaurant meal that reliably spikes the log. Seeing the pattern is usually enough to shift it. The history view is free, unlimited, and exports to CSV for your next clinic appointment.
Cited from the same sources
your endocrinologist reads.
- USDA
- ADA
- NIH
- ICMR
- NIN-Hyderabad
For diabetes management, the actionable number is glycemic load per meal — not calories, not a colour code. CalEye is the first patient-facing tool I've seen that surfaces that number from a photograph and shows you the source in the same breath.
Pricing
Get the app Free for the essentials. Pro for the long arc.
3-day free trial. Yearly billed annually. Cancel inside the app.
More on the diabetic angle.
All diabetes posts →
Every diabetes post in the CalEye archive.
48 cited, peer-reviewed posts on managing diabetes with photo-based tracking.
Carb counting fundamentals
- 20 Best Carb Counting Apps for Diabetes in 2026 Ranked by net-carb accuracy, CGM sync, restaurant database, and ease of use for both T1D and T2D management. Real research, real numbers, and a clear answer you
- Carb counting 101 — without the spreadsheet How to count carbs for diabetes management without a spreadsheet — using AI-driven photo recognition cited from USDA and ADA sources.
- Carb counting for Mediterranean breakfasts with diabetes Carb counting for Mediterranean breakfasts helps people with diabetes enjoy labneh, eggs, olives, and whole grain pita while keeping post-meal glucose in range.
- Carb counting for South Asian meals — chapati, rice, dal Carb counting for South Asian meals is notoriously difficult without reference data. Here are verified carb counts for chapati, rice, dal, and 20 other staples.
- How to Log a Buffet Meal — the 3-Photo Rule Buffet logging doesn't have to be guesswork. Use the 3-photo rule to capture your plate before, during, and after and let CalEye handle the math.
- Insulin-to-carb ratio explained for newly diagnosed Insulin-to-carb ratio tells you how much insulin covers each gram of carbohydrate you eat. Here's what it means, how it's set, and how to use it safely.
- Net Carbs vs Total Carbs: Which to Track — and When It Matters The fiber subtraction debate settled: which approach to use for weight loss, T2D management, and ketosis maintenance. The science, the numbers, and what actuall
- Pediatric diabetes — counting carbs for a school-aged child Pediatric diabetes carb counting requires parent-teacher coordination and age-appropriate targets. Here's the practical carb framework for children aged 5–12.
- Sugar alcohols and diabetes — the carb-counting gray zone Sugar alcohols are not calorie-free. Here's exactly how to count erythritol, xylitol, and maltitol in your daily carb budget when managing diabetes.
- Type 1 vs Type 2 diabetes — carb counting differences Type 1 vs Type 2 diabetes demand different carb counting strategies. Here's what the evidence says about precision, insulin ratios, and glycemic load.
- Why Fiber Doesn't Count as Carbs — The Net-Carb Derivation Net carbs subtract fiber from total carbohydrates, but the biochemistry is nuanced. Here is the science of digestibility, fermentation, and the regulatory math.
Glycemic load & index
- Glycemic load vs glycemic index — the one that matters Why glycemic load is the more useful metric for diabetic meal planning than glycemic index, and how AI photo analysis surfaces it.
- Postprandial Glucose Response — Individual Variability Research The same meal can spike glucose 2-3x differently across people. The science of inter-individual glycemic variability and what it means for nutrition apps.
- Reading a glucose curve — post-meal spike interpretation Reading your post-meal glucose curve tells you which foods your body handles and which push A1C up. Here's how to interpret the CGM data precisely.
- The Sydney GI Database — Methodology Explained The Sydney GI Database is the global gold standard for glycemic index values. Here is how foods are tested, averaged, and why values vary between labs.
CGM & monitoring
- A1C and what it actually means for daily eating A1C is the long arc of blood sugar — but the daily decisions that move it are smaller than most diabetics think. Here's the mechanism.
- CGM Accuracy — The Engineering Limits CGMs measure interstitial fluid, not blood, introducing physiological lag and calibration limits. Here is the accuracy science behind consumer glucose sensors.
- CGM vs A1C — when continuous glucose data overrides A1C Continuous glucose monitors can reveal what A1C hides: time-in-range, nocturnal lows, and post-meal spikes that drive complications even when A1C looks fine.
- HbA1c to eAG conversion — practical worked examples HbA1c to eAG conversion puts your three-month A1C into glucometer mg/dL units. Here are six worked examples across clinically relevant A1C values.
- Reading a glucose curve — post-meal spike interpretation Reading your post-meal glucose curve tells you which foods your body handles and which push A1C up. Here's how to interpret the CGM data precisely.
- Somogyi effect vs dawn phenomenon — getting the diagnosis right Somogyi effect and dawn phenomenon both cause high morning glucose but need opposite treatments. Here's the 3 AM test that tells them apart.
- The dawn phenomenon — why your fasting glucose spikes The dawn phenomenon causes fasting glucose to rise overnight without any food. Here's the hormonal mechanism, how to confirm it, and what to do about it.
Insulin, medications & emergencies
- Berberine vs Metformin for Blood Sugar: What the Trials Actually Show A clinical comparison of dosing, mechanism, side-effect profile, and glucose outcomes for berberine and metformin — cutting through the wellness noise to what.
- Diabetic ketoacidosis — recognizing the symptoms early Diabetic ketoacidosis symptoms begin subtly and escalate fast. Early DKA recognition — nausea, fruity breath, rapid breathing — prevents ICU admission.
- GLP-1 medications and meal tracking — what changes GLP-1 medications like Ozempic change how much you eat and how fast you digest food — and how you should track meals. Here's what actually shifts in practice.
- Hypoglycemia warning signs and meal-timing strategy Hypoglycemia warning signs can be subtle or absent. Knowing them and structuring meal timing correctly prevents dangerous low blood sugar episodes.
- Insulin-to-carb ratio explained for newly diagnosed Insulin-to-carb ratio tells you how much insulin covers each gram of carbohydrate you eat. Here's what it means, how it's set, and how to use it safely.
- Nutrition on GLP-1 Drugs: What to Eat When Ozempic Kills Your Appetite Protein floors, micronutrient density priorities, and meal-size strategies to avoid muscle loss on semaglutide — the nutritional framework GLP-1 prescribers.
- PCOS and Calorie Tracking: Why Standard Deficits Often Fail Insulin resistance, androgen-driven NEAT suppression, and the macro adjustments that actually move the needle for PCOS — beyond the generic 500 kcal deficit.
- The 15-15 rule for treating low blood sugar safely The 15-15 rule for low blood sugar is the ADA-recommended first-response protocol: 15g fast carbs, wait 15 minutes, recheck. Here's exactly how to apply it.
Complications & conditions
- Carb cycling for athletes with type 1 diabetes Carb cycling for type 1 diabetes athletes requires precise insulin adjustments and glucose monitoring. Here's the protocol used by competitive T1D athletes.
- Diabetic nephropathy — protein intake and kidney diet Diabetic nephropathy changes your protein requirements. Here's what the clinical evidence recommends for protein intake at each stage of kidney disease.
- Diabetic neuropathy — diet's role in slowing progression Diabetic neuropathy progression is linked to hyperglycemia and nutrient deficiencies. Here's what the evidence shows about diet's role in slowing nerve damage.
- Diabetic retinopathy — diet patterns linked to progression Diabetic retinopathy is modifiable with diet beyond glucose control. Here's the evidence on antioxidants, omega-3 fatty acids, and glycemic load.
- Gestational diabetes diet — the first 30 days A gestational diabetes diet in the first 30 days after diagnosis sets the trajectory for the rest of pregnancy. Here's exactly what to eat, avoid, and track.
- Gestational to type 2 diabetes — the post-pregnancy risk window Gestational diabetes raises your lifetime risk of type 2 diabetes 7-fold. The post-pregnancy risk window is real and preventable — here's the evidence.
- PCOS and Calorie Tracking: Why Standard Deficits Often Fail Insulin resistance, androgen-driven NEAT suppression, and the macro adjustments that actually move the needle for PCOS — beyond the generic 500 kcal deficit.
- Pediatric diabetes — counting carbs for a school-aged child Pediatric diabetes carb counting requires parent-teacher coordination and age-appropriate targets. Here's the practical carb framework for children aged 5–12.
- Perimenopause Weight Gain: The Hormonal Shifts Your App Isnt Oestrogen drop, visceral fat redistribution, and the protein and fibre targets that counter menopausal metabolic change — what the research says and what your.
- Pre-diabetes reversal — what the evidence actually shows Pre-diabetes reversal is achievable for many people, but the evidence is specific. Here's what the trials show about diet, exercise, and weight loss.
- Type 1 vs Type 2 diabetes — carb counting differences Type 1 vs Type 2 diabetes demand different carb counting strategies. Here's what the evidence says about precision, insulin ratios, and glycemic load.
- Type 2 diabetes and intermittent fasting — what the trials say Type 2 diabetes and intermittent fasting trials show real A1C reductions, but protocols differ. Here's what the evidence supports and what it doesn't.
Lifestyle, meals & travel
- Diabetes and Ramadan fasting — the clinical protocol Diabetes and Ramadan fasting can be managed safely with the right clinical protocol. Here's how to adjust insulin, meals, and monitoring across a 29-day fast.
- Diabetes and travel — managing meals across time zones Diabetes travel across time zones disrupts insulin timing and meals in predictable ways. Here's the clinical protocol for flying east, west, and long-haul.
- Glucose vs Fructose — Different Metabolic Pathways Glucose and fructose share a molecular formula but follow different metabolic routes. Here is the biochemistry, liver load, and why the sugar source matters.
- Hypoglycemia warning signs and meal-timing strategy Hypoglycemia warning signs can be subtle or absent. Knowing them and structuring meal timing correctly prevents dangerous low blood sugar episodes.
- Logging Meals When You're Sick Illness changes what and how much you eat. Here's the minimum viable CalEye tracking approach for sick days that keeps your habit chain intact.
- Resistant Starch — The Carb Your Gut Digests Differently Resistant starch escapes small-intestine digestion and feeds gut bacteria. The science of RS types, fermentation, short-chain fatty acids, and glycemic impact.
- The Microbiome's Role in Carb Metabolism Gut bacteria ferment undigested carbs and produce short-chain fatty acids that alter glycemic response. The science of microbiome-carbohydrate interactions.
- The PREDICT Studies — Personalized Nutrition Data PREDICT 1 and 2 are the largest personalized nutrition studies run. Here is the methodology, findings, and what they reveal about individual dietary response.
- Type 2 diabetes and intermittent fasting — what the trials say Type 2 diabetes and intermittent fasting trials show real A1C reductions, but protocols differ. Here's what the evidence supports and what it doesn't.
- USDA SR-Legacy — What's in the Database Your App Uses USDA SR-Legacy release 28 underpins almost every nutrition app. Here is what it contains, how values are measured, and where the gaps are for global cuisines.
Tools & apps
- 18 Best Apps for Prediabetes Management and Early Reversal From CGM pairing to carb-counting tools — the apps clinicians recommend when A1C sits between 5.7 and 6.4. The science, the numbers, and what actually works in
- Cronometer vs Carb Manager for Diabetes: Micronutrient Depth vs Keto Which app gives diabetics more useful daily data — Cronometer's micronutrient depth or Carb Manager's net-carb simplicity.
- MyNetDiary vs MyFitnessPal for Diabetes: Which Has Better Glucose Comparing A1C integration, carb-per-meal alerts, and CGM sync between the two most-used trackers for T2D. A clear, citation-backed answer with the practical num