FaceRating.ai Data Study
What's the Average Face Rating Score? We Analyzed 17,032 Faces
Based on 17,032 real analyses · January – August 2026 · Updated August 28, 2026
Data updated . Figures published before that date used the pre-December-2025 model scale and are superseded — the current figures on this page come from a single model on a single 1–10 scale, with no pooling across tools. These are model outputs from submitted photos, not a representative population survey or independent measurements of attractiveness. Associations between scores from the same model do not establish cause and effect.
Key findings
- • The average AI face rating is 5.59/10 (median 5.3).
- • Only 8.1% of people score a 7 or above, and just 0.2% score 9 or higher.
- • The single most common range is 5.0–5.9 (49.6% of everyone analyzed).
- • Eyes are the highest-scoring feature on average (5.97); the nose is the lowest (5.3).
- • Real faces hit only 77.28% golden-ratio compliance on average — the "perfect" ratio is almost never reached.
"What's a good face rating score?" is one of the most common questions people ask after using an AI face analyzer — and until now, most answers have been guesses. We looked at the actual data: 17,032 completed face-rating analyses from real users of FaceRating.ai, scored by the same AI model on the same 1–10 scale. Here's what the numbers actually say.
One thing to be clear about up front: these are the model's scores, not an objective verdict on anyone's worth. Attractiveness is subjective and cultural. But because every analysis uses the same model and the same scale, the distribution is a consistent, honest look at how the tool scores real people — which is exactly what you need to interpret your own result.
What do most people score out of 10?
Scores cluster hard in the middle. More than 4 in 10 people land in the 5.0–5.9 range, and the high end is genuinely rare — fewer than 1 in 200 analyses reach a 9.
| Score band | Analyses | Share | Distribution |
|---|---|---|---|
| Below 4.0 | 77 | 0.5% | |
| 4.0 – 4.9 | 4,230 | 24.8% | |
| 5.0 – 5.9 | 8,450 | 49.6% | |
| 6.0 – 6.9 | 2,891 | 17% | |
| 7.0 – 7.9 | 824 | 4.8% | |
| 8.0 – 8.9 | 524 | 3.1% | |
| 9.0 + | 36 | 0.2% |
Share of 17,032 analyses in each score band.
Is a 6 out of 10 a good face rating?
Because the scale bunches up in the middle, small score differences map to big percentile jumps. Here's where each score places you among everyone analyzed:
| If you scored… | You're roughly in the… |
|---|---|
| 5.0 or higher | top 74.7% |
| 6.0 or higher | top 25.1% |
| 7.0 or higher | top 8.1% |
| 8.0 or higher | top 3.3% |
| 9.0 or higher | top 0.2% |
So a 7 isn't average — it's genuinely rare. And the jump from 7 to 8 moves you from the top 8.1% to the top 3.3%.
How much can a face rating improve?
Alongside each current score, the model estimates a "potential" score — where someone could realistically land by improving the things they can control (grooming, skin, hair, styling). Across all 17,032 analyses, the average potential score is 6.9/10, an average gap of +1.31 points above the current score. Most of that gap lives in the most improvable features — which brings us to the feature breakdown.
Which facial feature scores highest?
Averaged across everyone, eyes and eyebrows score highest, while skin and nose score lowest. That's encouraging: the weakest-scoring areas (skin especially) are also the most improvable without anything drastic.
| Feature | Mean score | |
|---|---|---|
| Eyes | 5.97 | |
| Eyebrows | 5.94 | |
| Lips | 5.65 | |
| Symmetry | 5.62 | |
| Bone structure | 5.6 | |
| Jaw | 5.59 | |
| Facial harmony | 5.58 | |
| Cheeks | 5.42 | |
| Chin | 5.39 | |
| Forehead | 5.39 | |
| Nose | 5.36 | |
| Skin | 5.3 |
Mean feature score (0–10) across all analyses.
The golden ratio is (mostly) a myth
The "golden ratio" (1.618) is often sold as the secret formula of a perfect face. In the data, real faces reach only 77.28% compliance with golden-ratio proportions on average, and even the highest-scoring faces top out around 96%. Facial symmetry tells the same story: the average symmetry score is 5.62/10, and only 3.3% of people score 8 or above. Near-perfect symmetry is the exception, not the standard — which is worth remembering the next time an app implies you're falling short of an "ideal."
The most common face shapes
| Face shape | Share | |
|---|---|---|
| Oblong | 44.4% | |
| Oval | 42.3% | |
| Round | 6.7% | |
| Other / hybrid | 4.1% | |
| Square | 2.6% |
Share of analyses by detected face shape.
Oval and oblong dominate, together making up around 86% of faces. Face shape drives a lot of practical advice — which is why our hairstyle recommendations are matched to it.
Methodology
This study aggregates 17,032 completed face-rating analyses performed on FaceRating.ai between January – August 2026. Each analysis is produced by the same AI vision model on a consistent 1–10 scale, with feature-level sub-scores, a golden-ratio compliance estimate, and a detected face shape. All figures are aggregate and anonymized — no individual results, images, or personal data are included or identifiable. Scores reflect the model's assessment and are not an objective measure of attractiveness, which is inherently subjective. Percentile figures are the share of analyses at or above each score.
See where you land
Get your own AI face rating, feature breakdown, and potential score — free, no app to download.
Keep reading
Cite this study
These figures are ours, from our own data, and you are welcome to quote them in articles, videos or research with attribution and a link.
FaceRating.ai (2026). What's the Average Face Rating Score? We Analyzed 17,032 Faces. Analysis of 17,032 AI facial analyses, January - August 2026. https://facerating.ai/research/average-face-rating-scoreHow the data was collected
- 17,032 analyses run on FaceRating.ai during January - August 2026. Aggregate and anonymised; no individual result is published, and no image is included in any figure.
- Scores are produced by a vision model on a fixed 1-10 scale, not by human raters. They describe how this model scores faces, which is what a rating tool is actually asked.
- The sample is self-selected. These are people who chose to have their face rated, so it is not a random sample of any population, and the distribution should not be read as one.
Writing something and want a figure we have not published, or a check on a number before you print it? Ask us.