Skip to content

Press

BecomeTen press kit

BecomeTen is an AI facial-rating app (web, iOS, Android) from Slovakia. One photo returns a strict score, a trait-by-trait breakdown and a realistic estimate of what can change without surgery. Every step carries an evidence level from a curated, PubMed-verified library. About thirty free face-analysis tools run in the browser and never upload the photo.

Read the evidence

A claim should come with its context.

Strong
Moderate
Weak
No studies

Evidence levels describe support for a claim, not a guaranteed result.

Read the methodology

Downloads

Live demo of the open-source tool math: martinsaraka.github.io/face-metrics

Listed on

Boilerplate

Short: BecomeTen rates your face from one photo and tells you, honestly, what you can change. Free on-device tools, paid full report, every tip labelled with its evidence level.

Long: BecomeTen is an AI facial-rating app that gives honest numbers instead of flattery. Upload one photo and you get an overall score on a 0–100 scale calibrated so an average, groomed adult lands in the middle, a breakdown across five areas and the visible traits, and a capped estimate of how much of that can move without surgery. From there it builds a personal plan in which every recommendation carries an evidence level and may cite only a curated, PubMed-verified evidence library; steps without studies are labelled as such. More than twenty free tools, from face shape and symmetry to eye type and fWHR, run on-device with MediaPipe's face mesh, so the photo never leaves the browser. BecomeTen is built by Appinara s. r. o. in Slovakia and launched in 2026.

  • Product: web app at becometen.com, English only, adults only. Also an iPhone app on the App Store and an Android app on Google Play.
  • Company: Appinara s. r. o., a limited company registered in Slovakia, operating under EU law. Registered address and identifiers are on the imprint.
  • Founder: Martin Saraka, software engineer. Not a clinician, and the product says so on every relevant page. Bio and profiles on the about page.
  • Pricing: the first daily web score, tier and share card are free. A full report is a one-time €6.99; PRO is €9.99 a month or €59.99 a year for 20 scans a month. Direct full reports, extra daily scans and in-app reports require payment. Details on the pricing page.
  • Privacy: a free scan's photo is processed and deleted; no photo is used to train a model. The on-device tools send nothing at all. The privacy policy lists every processor.
  • Evidence: the plan may only cite from a hand-curated library whose entries were resolved through PubMed's own lookup. Unsupported steps are labelled as having no studies. The rules are on the methodology page.
  • Open source: the geometry behind the free tools is published on GitHub as face-metrics (MartinSaraka/face-metrics, MIT licence), with a live demo, so anyone can audit how a face shape or symmetry score is derived.

A face rater has an incentive to flatter, because a flattering result gets shared and an honest one does not. BecomeTen is calibrated the other way: scores above 80 are reserved for genuinely striking structure, and every area has a hard cap on non-surgical gain that the server enforces in code, whatever the model returns.

The second difference is the evidence gate. A language model asked for a citation will sometimes produce a plausible reference that does not exist. BecomeTen closes that door by letting the model cite only from a fixed, verified library, and by checking every citation server-side before it reaches the reader. The third is the on-device tool set: face shape, symmetry, golden ratio, canthal tilt, jawline, facial thirds and fifths, eye shape and pupillary distance, all computed from landmarks in the browser with no upload. The full list is on the tools page.

  • It does not report measurements from a photo in the rating. Structural traits come back as directional bands; only the free single-measurement tools show a labelled photo estimate in numbers.
  • It does not promise anything about bone, and it never recommends surgery or anything that damages tissue.
  • It does not rank users against each other, and it does not adjust the score for ethnicity or gender identity.
  • It does not use the looksmaxxing community's demeaning labels; its tiers are neutral bands on its own scale.

The download links above this section include the logo in PNG and SVG, the store feature graphic, seven phone screenshots at 1080 by 1920 pixels, and a single zip with everything. Screenshots use AI-generated demo faces, never real users. Please credit BecomeTen and link to becometen.com. Do not present the score as a medical or diagnostic result.

For interviews, review access, a free full report for testing or a walkthrough of the scoring rubric, write to hello@becometen.com. Messages are read and answered by the founder, usually within a day. Comparisons with other apps are welcome; the compare pages show how BecomeTen positions itself against each one.

Sources

  1. 1.Sunscreen and prevention of skin aging: a randomized trial — Annals of Internal Medicine (2013)(Opens in a new window)
About this content
An estimate, not a measurement. Educational content, not a diagnosis or a substitute for a doctor's consultation.
Methodology & evidence