YOUR TYPE, IN PICTURES

MyType FYICompare my type with friends

Compare my type with friends through separate tests and an explicit public card. Compatible tests can compare model-ranked portraits in the same collection, with the shared count and denominator shown.

10 rounds free · No selfie needed

?

Tap any face you like: it is saved and replaced. Pick as many as you want, then move on. Your first 10 rounds and base result are free. Undo reverses your latest pick.

Round 1 / 500 picks
Prediction confidence: 0.0%

First 10 rounds free · then 1 Credit / 5 rounds

By starting, you agree to our terms and privacy policy.

A CLOSER LOOK

About this face preference test

Published · Independent guide and face preference test

Compare my type with friends after separate tests

To compare my type with friends, each person takes their own free test first. Avoid choosing on someone else’s behalf or copying their selections to make the results agree. You can then publish a card containing selected fictional portraits and visual patterns. Use the pictures to ask what each person noticed, especially when you expected similar preferences and discovered different choices instead.

Compare my type with friends using a public card

The compare my type with friends flow uses an explicit public snapshot rather than exposing a private account. Use the explicit share action from your result. The public card contains a limited snapshot and excludes your email and account details. Anyone with its link can read it until expiry or deletion. Sending your private result URL is not the same as creating a share card.

Compare my type with friends through ranked overlap

When I compare my type with friends, the numerical result concerns model-ranked portraits in the same collection. Open a friend’s shared link and complete your own 20-crowd test. For compatible tests, we compare their top-ranked quarter of the same portrait collection. The report shows the shared count and denominator. This experimental ranking comparison is not a measured probability that you like the same real people.

Compare my type with friends across matching portrait groups

The compare my type with friends calculation requires compatible portrait groups and collection versions. A women-only test and a men-only test do not rank the same candidate pool. A mismatch should be explained instead of producing a convincing-looking percentage from unrelated examples.

Compare my type with friends without sharing account access

To compare my type with friends safely, use the share control rather than exchanging passwords or a private account link. The public card contains selected faces, visible patterns and a count. Anyone with the link can read that snapshot until expiry or deletion.

Compare my type with friends as a conversation

I can compare my type with friends without treating agreement as a relationship test. Ask which full portraits appealed, what details each person noticed, and where the summary feels incomplete. Shared visual taste does not establish emotional compatibility or predict real dating choices.

Sources & scope

For the original service, see MyType’s own explanation and current terms. Our test at mytypefyi.app has its own features, data handling and pricing. Features and prices at other services can change.

GOOD TO KNOW

FAQ: a few things you might wonder.

Can I compare my type with friends without sharing my account?

Yes. To compare my type with friends, publish a limited card and share that link. Your email, password and private account access are not included.

Can I compare my type with friends in different portrait groups?

The compare my type with friends calculation requires compatible candidate pools. Different presentation groups or collection versions should show a mismatch, not a fabricated overlap number.

Does compare my type with friends measure romantic compatibility?

No. To compare my type with friends, this version compares the direction of your two visible-preference weight vectors using cosine similarity. It does not measure relationship compatibility or real-world dating probability. A comparison stays unavailable when either test has too little signal.