Processing-first scanning

Look past the healthy-looking front label.

Use Lychee to inspect processing, ingredients and nutrition together. Its food score puts processing first, with NOVA data and ingredient-list evidence when available.

Free download · Optional Premium · iOS & Android

NOVA is a useful input, not the whole verdict.

NOVA groups foods by the nature and purpose of processing. Lychee uses available NOVA information and can estimate processing from ingredient signals when database information is absent and the evidence supports an estimate.

Ingredient names, formulations and product categories matter. The model has category-specific checks and can use label evidence to challenge an inconsistent catalog tag. Read the food scoring methodology for the approach.

Start with the food you already buy.

Breakfast cereal

Compare the ingredient lists and processing signals of cereals used for a similar breakfast. Review sugar and fiber too.

Snacks and bars

Look beyond protein, low-sugar and vitamin claims. Read the formulation, sweeteners and additive information.

Sauces and pantry staples

Check the ingredients and available nutrition, save what you buy and compare alternatives over time.

Ask for evidence before a number.

A missing ingredient list or nutrition field should reduce confidence. Lychee can withhold a food score when information is insufficient. A long label alone is not proof of harm, and a short one does not resolve allergies or portion context.

Before you scan.

How does Lychee score food?

Lychee uses a 1–100 food score that prioritizes processing. Additives, ingredient quality, sugar, salt, saturated fat, fiber and protein also matter. Available NOVA data and ingredient-based estimates inform processing; Lychee applies its own scoring rules. Missing data can reduce confidence or prevent a score.

What if my product is missing from the database?

Try reading the ingredient label with Lychee. You can also submit or correct product data through its contribution flow. OCR and contributed records need checking; a missing ingredient list should be treated as incomplete data.

More detail, when you need it.

Start with your next label.

Food and beauty ingredient context in your pocket.