Deviation

The IQ test that shows its work

Five kinds of thinking, fifteen minutes, one score you can actually interpret. No sign-up, no email, no paywall, no results held hostage.

~15 minutes30 items + a timed roundNothing to sign up forFree, no upsell

What you get at the end

A score with a range

Not one flattering number. You get the interval your answers actually support, so you know how much to trust it.

A profile, not a total

All five areas scored separately. The shape is usually more interesting than the overall height.

A link that just works

Your whole result travels inside its own URL. Share it, bookmark it, come back to it — there is no account and no server holding it.

What you’ll do

Five rounds, one after another, each with its own clock. You can stop between rounds but not inside one.

1

Fluid reasoning

8 items · up to 8 min

Complete the pattern. Each grid follows a rule — find it, then pick the piece that belongs in the empty cell.

2

Quantitative reasoning

6 items · up to 5 min

Each sequence follows a rule. Work out the next number and type it in.

3

Verbal reasoning

6 items · up to 4 min

Work out how the first pair relates, then complete the second pair the same way.

4

Working memory

10 items · up to 6 min

Watch the digits, then type them back in reverse order.

5

Processing speed

60 seconds

Match each symbol to its digit using the key. As many as you can in 60 seconds.

Why you can trust the number

Most free IQ tests either hide their method or don’t have one. This one publishes the whole thing: how each round is scored, how the scores combine, where the reference figures come from, and — the part nobody else writes down — exactly which parts are estimates rather than measurements.

It also refuses to flatter you. Scores are capped at the range a fifteen-minute test can honestly resolve, and every result is shown with its margin of error rather than as a single confident digit.

Read the methodology
One thing to be clear about. This is not a clinical assessment and is not suitable for diagnosis, educational placement, or hiring decisions. It scores against a constructed reference distribution rather than a measured population, and is built for adults.