NO SIGN-UP REQUIRED

FREE AIattractiveness test— NO SIGN-UP

Upload one photo and get an instant, science-based attractiveness score. Our VisionScale™ engine measures the facial proportions, symmetry and structural traits that decades of published research link to attractiveness — and tells you exactly how it reached the number.

142
Facial Landmarks Mapped
99.8%
Landmark Detection Rate
100%
Photos Never Stored

Free forever

No paywall, no premium tier

No sign-up

No account, no email required

Photos never stored

Analysed in the moment, then discarded

Results in seconds

Full breakdown, not just a number

How the attractiveness test works

1

Upload a photo.

A clear, front-facing photo works best — camera at eye level, even lighting, neutral expression. Close-up selfies distort facial proportions, so step back if you can.

2

We map your facial landmarks.

The engine detects the reference points that define your facial geometry — eye corners, nose base, mouth edges, jaw contour, hairline — and computes the proportional relationships between them.

3

You get a score and a full breakdown.

Not just a number: a percentile against population norms, plus component scores for proportional harmony, symmetry and structural traits, so you can see what's driving the result.

Want the most accurate reading? Our guide to taking the right photo covers the distance, angle and lighting that make the difference — camera distance alone measurably changes facial proportions.

What the test measures — and why

Most face rating tools hand you a number and no explanation. Here's exactly what ours is computing, and the research behind each component.

Proportional harmony.

How closely your facial proportions sit to the arrangements that maximise attractiveness ratings in controlled studies. Research published in Vision Research identified specific optima — eye-to-mouth distance at roughly 36% of face length, interocular distance at roughly 46% of face width — and found that these match the proportions of an average face.

Facial symmetry.

The degree of bilateral correspondence between the two halves of your face, measured landmark by landmark and normalised for face size. Symmetry is one of three traits that meta-analytic reviews confirm as attractive in both male and female faces and across cultures.

Structural and dimorphic traits.

Jaw definition, cheekbone position, brow structure and the other features that contribute to how a face reads as a whole. Faces are processed holistically, so these are weighted in configuration rather than scored in isolation.

Skin quality.

Texture and tone, which the research literature identifies as an independent contributor to face preference alongside shape-based traits — and the most modifiable of everything we measure.

For the full picture, read what makes a face attractive or see our complete methodology.

What a score can and can't tell you

We'd rather explain this than have you find out and stop trusting the result.

Your score estimates the shared, structural layer of facial attractiveness — how a panel of raters would likely score this particular photo, based on measurable traits. That's a real signal: research consistently finds substantial agreement about facial attractiveness both within and across cultures.

What no test can measure is the individual layer. A twin study published in Current Biology found that most reliable variation in personal face preference traces to each person's unique environment and life experience — the faces they've encountered, their particular social history. That information isn't in your face, so it can't be in your score.

So: treat your result as one data point about structural convention. It is not a verdict on your appeal, your prospects, or your worth. Anyone claiming to predict whether a specific person will find you attractive is selling something that doesn't exist.

We break this down in full in how accurate are AI attractiveness tests.

Your photo is never stored

Your photo is processed for analysis and then discarded. We don't save it, publish it, use it in a gallery, or use it for training or marketing. There's no account to create, so there's nothing tied to your identity in the first place.

Full details in our privacy policy.

Frequently asked questions

Is the attractiveness test free?
Yes — completely free, with no sign-up, no account, and no premium tier. You don't need to provide an email address.
How accurate is the attractiveness test?
It's accurate at a specific task: estimating how a panel of human raters would score the photo you upload, based on structural traits with established links to attractiveness ratings. It can't predict whether a specific individual will find you attractive, because research shows that depends primarily on the viewer's own life experience rather than on your face.
Do you store my photo?
No. Your photo is analysed and then discarded. It's never saved, shared, published, or used for training.
How attractive am I? Can this test really tell me?
It can tell you how your facial proportions, symmetry and structural traits compare to population norms, and how a group of raters would likely score that image. That's genuinely informative — and narrower than a verdict on your appeal. Read what a score means for the full interpretation.
Why did my score change with a different photo?
Because photos differ in how faithfully they capture facial geometry. Camera distance is the biggest factor — a close selfie measurably widens the apparent nasal base compared with a photo taken from several feet away. Lighting, head angle and expression add further variation.
What photo works best?
Front-facing, neutral expression, even lighting, camera at eye level, taken from at least four feet away. Use the rear camera and crop afterwards rather than holding the phone close.
Do I need to create an account?
No. There's no sign-up step at any point.
Is this the same as a PSL rating?
No. PSL is a scale developed by online looksmaxxing communities with a specific ideological framing. We report percentiles against population norms and explain our components. Read our breakdown of the PSL scale.