# Do we look alike?

> Free look-alike test. Upload two photos and see how alike the two faces are, with a similarity score and a plain-English reading. No sign-up, never stored.

Source: https://sightradar.com/tools/do-we-look-alike

Add a photo of each person and see how much the two faces have in common, as a score from 0 to 1 with a plain-English reading.

Use it in the browser: https://sightradar.com/tools/do-we-look-alike (no sign-up). Images are processed in memory and are not stored. No photo, face crop or faceprint is saved, and nothing is used for training.

## What the result means

The score is cosine similarity between two face embeddings, from 0 to 1. The page reads it as resemblance, against the default same-person line of 0.42:

| Score | Reading |
| --- | --- |
| 0.42 and above | Easily mistaken for each other (the API would call the two photos the same person) |
| 0.28 to 0.42 | A strong resemblance |
| 0.15 to 0.28 | Some resemblance |
| below 0.15 | Not much resemblance |

The bands are this page's own reading of the score. The score measures how alike two faces look. It does not say whether two people are related, and it is not a DNA test.

When a photo has several faces, the largest, most confident one is used.

## Paid API

`POST https://api.sightradar.com/v1/compare` (Rekognition CompareFaces equivalent) costs 186 credits for two images (93 per image). Reference: https://sightradar.com/docs/api/faces/compare-two-faces.md

## Free API for AI agents and scripts

AI agents, scripts and quick tests can call this tool directly over HTTP. Free to test under a fair-use policy, no key needed. For regular or production use, sign up and use an API key: there is no free-tier cap on a key, and usage is billed per photo on prepaid credits.

```bash
curl -X POST https://sightradar.com/api/free/compare -F "source=@a.jpg" -F "target=@b.jpg"
```

- Body: `multipart/form-data`, fields `source` and `target`. JPG, PNG or WebP, up to 10 MB each.
- Response: JSON `{"data": {...}}`. Images are processed in memory and never stored.
- `GET https://sightradar.com/api/free/compare` returns a JSON description of the endpoint.
- If the free fair-use limit is reached, the API answers `429` with `"code": "free_limit_reached"`, a `signup_url` and `how_to_continue`.

To use it as much as you need: sign up at https://sightradar.com, create an API key in the console, and call `POST https://api.sightradar.com/v1/compare` with `Authorization: Bearer <api-key>` and a JSON body `{"sourceUrl": "...", "targetUrl": "..."}` (the paid compare takes image URLs).

Full contract: https://sightradar.com/docs/free-api.md

## FAQ

### How does the look-alike test work?

Each face is turned into an embedding, a list of numbers that summarises its features, and the two embeddings are compared with cosine similarity. The result runs from 0 to 1. The closer to 1, the more the two faces have in common. It is the same measurement face recognition uses to decide whether two photos show one person.

### What do the results mean?

The tool reads the score against the line the API uses to call two photos the same person. Above that line, the two could easily be mistaken for each other. Just under it is a strong resemblance, further down is some resemblance, and a low score means the faces have little in common. Most pairs of different people score well under the line, so a strong resemblance is unusual.

### Is the score a percentage?

Not quite. Some tools show the same number as a percent, so 0.37 becomes 37%, but it is not the share of features two people have in common. It is a measure of how close two faces are as the model sees them, on a scale from 0 to 1.

### Can it tell if two people are related?

No. Relatives can score higher than strangers, and so can unrelated people who happen to look alike. A score can't tell those apart, and it is not a DNA test. Read it as how alike two faces look, and nothing more.

### Which photos work best?

One clear face per photo, large in the frame, well lit and facing the camera. Photos taken at a similar age and angle give the fairest comparison. Sunglasses, heavy filters and side-on shots lower the score. When a photo has several faces, the largest, most confident one is used.

### Are my photos stored?

No. Both photos are processed in memory to work out the score and then discarded. No photo, face crop or faceprint is saved, and nothing is used to train models. The picture you can download of the result is made in your browser.

### Can I use anyone's photo?

Use photos of yourself and of people who are happy to be compared. The tool does not identify anyone or search for a person: it only compares the two photos you give it.

### Is the look-alike test free?

Yes. There is no sign-up and no card. A fair-use limit per visitor keeps the tool fast for everyone; if you reach it, the tool shows how long to wait.
