Yes — AI can estimate the calories and macros in a meal from a photo, and it's genuinely useful for fast logging. But it's an estimate, not a lab measurement. You snap your plate, the AI recognises the foods and portions, and returns calories, protein, carbs and fat in seconds. Here's how it works and how much to trust it.
How it works
A vision model trained on huge numbers of food images identifies what's on your plate, estimates portion sizes, and matches each item to nutrition data to produce calories and macros. Full explainer: how an AI calorie counter works.
How accurate is it?
Good for everyday tracking — typically in the right ballpark — but not exact. It can't see:
- Hidden oil, butter or sauces cooked into the dish.
- Exact density of what's underneath or mixed in.
- Precise portion weight from a single angle.
So treat it as a fast, solid estimate. For tracking, a consistent estimate you keep up beats a perfect one you abandon (how accurate are calorie apps).
Where it shines vs where to use a barcode
- Photo AI is best for meals — home-cooked, restaurant, anything unpackaged (how to track calories with a photo).
- Barcode scanning is best for packaged food, since it pulls the exact label.
The best apps do both, so you use whichever is faster.
Get better estimates
Good light, a top-down angle, one plate at a time, and confirm the items the AI detects — especially logging any oils.
Snap a meal, get the numbers
forme estimates calories and macros from a photo of your meal, then scores it for your goals — and scans barcodes too, whichever's quicker.
The bottom line
AI can count calories from a photo and it's great for fast logging — just treat the numbers as a good estimate, log hidden oils, and use barcodes for packaged food. This is general information, not medical or dietary advice.