Any digital scale
Use the scale you already own. The number on its display is the weight, so there is nothing to pair and no new scale to buy.
You eat.
It logs.
Logging food means typing every meal, and the typing kills the habit.
Wear Ray-Ban Meta glasses, put the food on any digital scale, and it goes to Apple Health. Hands free.
How it works
No manual entry. Hands free.
Typing apps make you enter every food. Photo apps guess the portion from a picture. itlogs.food reads the number your scale already shows.
Ray-Ban Meta glasses
The glasses see the food and the display. Muse Spark names the food and reads the grams. For a packaged food, it reads the wrapper and searches the web for the label. For a dish it does not know, it finds typical values.
Put the food down, hands free
No phone in your hand, no photo to take, nothing to type. Move on to the next food.
It goes to Apple Health
The food, the grams, the calories and the macros. You check the meal, then send it with one tap.
Compare
Weighed, not guessed.
| itlogs.food | Qalzy | Photo apps | Typing apps | Smart scales | |
|---|---|---|---|---|---|
| What you need | Ray-Ban Meta glasses, any digital scale, an iPhone and, today, a Mac | Its own scale | A phone | A phone | Its own scale and a phone |
| Weighs, does not estimate | |||||
| Names the food with AI | |||||
| Weighs and names in one step | |||||
| Works with the scale you already own | |||||
| Hands free while weighing | |||||
| No phone in your hand while weighing | |||||
| Goes to Apple Health | |||||
| Names packaged food and finds its label |
yes partly or it varies no
In itlogs.food you tap Start, Stop and send on the phone, and nothing for each food. It reads a packaged food's wrapper, and the first lookup of a product takes 30 to 60 s. Qalzy's column comes from its own guide, read 2026-NOVEMBER-04: you press a button on the scale for each scan, or twice for a Live Scan.
Four popular photo apps logged calories and fat about one-third too low, across about 100 meals measured to 0.1 g. Study led by an NIH fellow, presented at the American Society for Nutrition's annual meeting.
The log
A log that writes itself.
Every line carries the grams, the calories and protein from a nutrient table. You check the list, then one tap sends it to Apple Health.
- Grams to the unit, straight off the display
- Protein, carbs, fat and kcal from the grams
- Sent to Apple Health with one tap
- 20:15apple112 g58 kcallogged
- 20:15banana124 g110 kcallogged
- 20:16
total236 g169 kcal1.6 g protein
In the kitchen
Glasses on. Scale on. That's the setup.
“Every other health stream fills itself: the ring, the watch, the body scale. Food was the one log still typed by hand. The typing is what kills the habit.”
Pictures generated with Muse Image, Meta's image model.
Built on
Meta's models, in the loop.
Ray-Ban MetaThe glasses, through the Wearables Device Access Toolkit: the camera stream, and speech recognized on the glasses.
Muse SparkReads the frame: the food, the number on the display, and how sure it is, as structured output. For a packaged food, it reads the wrapper, then its web search finds the printed numbers.
Muse Voice TranscribeThe fallback for speech when the glasses are not recognizing it. Realtime, 24 kHz.
Muse ImageMade the kitchen pictures on this page from a text prompt and one reference frame.
Muse Spark OpenThe open-weight path: the same reader on Baseten, one setting away. It read the apple right at 20 s a read, too slow for live use, so Muse Spark runs the demo.
FastAPI + SwiftA thin phone client streams frames up a WebSocket; one Python loop does the rest. Open source under Apache 2.0.
Built for the Meta Global AI
Developer Hackathon.
The code, every recorded run, and a replay of real sessions you can run with no glasses and no key.