Best food scanner apps: five compared on what they read
A food scanner app reads a barcode, looks the product up in a database and prints a verdict. The five worth comparing in the US — Yuka, Open Food Facts, Bobby Approved, Fig and LabelPeek — differ hardly at all in the scanning, and almost entirely in the database and the scoring rule sitting behind it.
This page sets out what each one reads, whose data it runs on, how it is sold and who it suits. LabelPeek is included and listed last, with its own limits stated first: the app is not released, and the web checkers are what exists today.
What actually separates one food scanner from another
Food scanner apps differ in three places: the product database they query, the rule that turns an ingredient list into a score, and what they do when a barcode is missing from the database. The camera is the same commodity component in every one of them, and it is the least interesting part of the comparison.
The database
Coverage decides whether the scan works at all. A US private-label product missing from the database returns nothing, whatever the rule would have said.
The rule
Every score is a weighting someone chose. Move the weight from nutrition to additives and the same product changes colour without a single fact changing.
The miss
What the app does with an unknown entry matters more than the hit rate. Printing a clean verdict over a gap in the data is the failure worth checking for.
No regulator scores packaged food. Nutri-Score is a public labelling scheme, but the ranked verdicts these apps print are private compilations, and two of them can disagree about one product without either being wrong on the facts. What is verifiable is the regulatory status of individual additives, and that covers a minority of a label.
25 of 78
Red 3 is one of the 25: the FDA revoked the food authorisation on 15 January 2025, with the ban taking effect 15 January 2027 for food and 18 January 2028 for ingested drugs (90 FR 4628; 21 CFR 74.303). Titanium dioxide is another — removed from the EU list of permitted food additives by Commission Regulation (EU) 2022/63, effective 7 February 2022, and still permitted in the US at up to 1% by weight of the food (21 CFR 73.575). Everything outside that small set is judgement.
The five food scanner apps side by side
Cost below is given as a model rather than a price. Tiers move without notice, store listings are the only current source, and a dollar figure copied into an article ages badly within a quarter. The columns that matter for a choice are the data source and the rule, because those change what the app tells you.
| App | What it reads | Data behind the verdict | Cost model | Suits |
|---|---|---|---|---|
| Yuka | Barcodes, food and cosmetics | Own product database, published weighting across nutrition, additives and organic status | Free to scan, optional paid tier | Shoppers who want one number per product |
| Open Food Facts | Barcodes, food | Volunteer-contributed open database, ODbL 1.0, edit history visible | Free, non-profit, no paid tier | People who want the raw record instead of a score |
| Bobby Approved | Barcodes, food | Criteria list written by Bobby Parrish of FlavCity | Free to download (US App Store, checked 11 August 2026) | Shoppers who want one consistent editorial standard |
| Fig | Barcodes, food | Packaged-grocery database filtered by the diet you configure | Free tier plus paid upgrade | Allergies, low-FODMAP and other specific protocols |
| LabelPeek | Pasted ingredient text, no barcode | Hand-written reference of 78 entries, 74 of them markers of industrial processing, 25 with a named regulatory source | Web checkers free, app not released | Reading a label you already have in front of you |
What each app does in detail
Each entry below covers the same four things: what the app reads, what data stands behind the answer, how it is sold and the shopper it fits. They are ordered by how widely they are used in the US, not by preference, and LabelPeek is last because it is the newest and the least finished of the five.
Yuka
Yuka scans a barcode and returns a score out of 100 with a colour band, covering food and cosmetics in one app. The food score is weighted: most of it from nutritional quality on the Nutri-Score method, a smaller share from additives, and a bonus for organic certification. Scanning is free, and a paid tier adds offline use and search without a barcode. That weighting is Yuka’s own editorial choice, published on its site rather than issued by any authority, which is why a product can rate well there and poorly under a rule leaning harder on additives. If the score is what you are questioning rather than the app, alternatives to Yuka compared sets its rule against three others on the same criteria.
Open Food Facts
Open Food Facts is a collaborative product database rather than a verdict engine, run by a French non-profit and published under the Open Database License 1.0. The app scans a barcode and shows what contributors have entered — ingredient list, Nutri-Score, NOVA processing group, additives — with the edit history and the photo of the packet attached. It is free, carries no advertising and has no paid tier, and it is the upstream source several commercial scanners draw on. The trade-off is coverage: entries are volunteer-contributed, so a US private-label product may be missing, partial or out of date. It is also the only one of the five that shows its own gaps rather than filling them in.
Bobby Approved
Bobby Approved is the FlavCity food scanner built by Bobby Parrish, and it needs disambiguating before anything else: an unrelated subscription-tracking product also trades under the name Bobby, and search results mix the two. The app scans a barcode and returns an approved or not-approved answer against a criteria list its author wrote, covering seed oils, certain sweeteners, synthetic dyes and preservatives among others. That is the design rather than a flaw — one person’s standard applied consistently, which is what some shoppers want and others find too blunt for a weekly shop. It is free to download on the US App Store as of 11 August 2026; check the listing for in-app purchases. Alternatives to Bobby Approved covers what changes when the criteria list is published in full instead of held privately.
Fig
Fig inverts the universal score. You tell it what you are avoiding — a named allergen, a low-FODMAP protocol, added sugar, a religious restriction — and the scan answers that question only. A product failing your filter is flagged with the reason you set, and the same product shown to somebody with a different profile passes without comment. The database covers packaged grocery items, and the app runs a free tier with a paid upgrade for more filters at once. Fig is therefore the most useful of the five when the question is dietary and specific, and the least useful when there is no filter to set: configure nothing and Fig has nothing to say.
LabelPeek
LabelPeek has no app in the stores yet, and any page claiming otherwise is wrong. What exists today is a pair of web checkers that take pasted text rather than a barcode. The ultra-processed food checker matches an ingredient list against a hand-written reference of 78 entries, 25 of which carry a regulatory restriction with the source document named. Its output has three states instead of two — matched, recognised but not on the list, and not recognised at all — so a gap in the reference is reported as a gap rather than printed as a clean result. Nothing is uploaded, there is no account, and camera reading is deliberately held back for the app rather than attempted in the browser.
When a food scanner app is the wrong tool
A food scanner app is the wrong tool when the food has no barcode. Anyone cooking from raw ingredients — produce, meat, dry beans, flour — is buying items with nothing to scan and no ingredient list to read, and a scanner adds nothing to a basket that was never processed in the first place.
Three other cases fall outside what any of the five can do. Restaurant and deli food carries no label, so the scan has no input. Portion size is invisible to a barcode: the same code returns the same answer whether you eat a spoonful or the packet. And none of these apps is a diagnostic instrument — a scanner is a screening tool, not a diagnosis or a substitute for medical care.
A scanner is a shopping tool, not a diet. It compares two boxes on a shelf. It has nothing to say about the meal you cook from things that never came in a box.
Apps covering cosmetics as well as food run that half on a different kind of list, worth knowing before trusting one score across both. No regulator anywhere issues a comedogenic list; every published one is a private compilation, and they contradict each other on the same ingredient. The best ingredient checker apps comparison covers the tools handling both sides and what each list is made of.
Where LabelPeek is the wrong choice
LabelPeek is the wrong choice in the supermarket aisle. There is no released app, no camera and no barcode lookup, so comparing two boxes on a shelf would mean typing an ingredient list into a phone, which nobody does. A barcode scanner wins that job outright.
LabelPeek web checkers
Paste an ingredient list and get a three-state result matched against a reference published in full on the same page. Runs in the browser, no account, nothing uploaded.
A barcode scanner such as Yuka
Point the camera at a barcode and get an answer in about a second, from a database covering millions of packaged products.
It is also the wrong choice for counting calories or macros, which it does not do, and for allergen decisions, where a purpose-built filter beats a general additive reference. The narrow case it fits is a label you can already see, checked against a list you can read yourself.
If the question is narrower than “which scanner”, two pages answer it more directly: apps that scan food ingredients covers the job rather than the products, and alternatives to Open Food Facts separates people replacing the shopping app from people replacing the open dataset.