IQTest EN

Comparison

RIOT IQ vs iq-test.io: which free IQ test is worth your ten minutes?

The most rigorous test in the comparison, and the only one that costs real money.

Updated September 2026 Reviewed by the IQTest EN editorial team 9 min read

The short verdict

If you want a free IQ test where the full result is genuinely free, iq-test.io is the one we point people to. Thirty-three questions, no timer, no email gate, and a methodology page that describes the scoring model rather than asserting accuracy.

RIOT IQ is the better pick in one specific case: you want the closest thing to a real psychometric instrument without booking a psychologist. Its real strength is that built on the cattell–horn–carroll model by an intelligence researcher, with a published test manual. Its weakness is that the price puts it in a different category from every other test here, and it is still not a clinical administration.

Side by side

iq-test.ioRIOT IQ
Length33 questions, no timer, about 10 minutes15 subtests across six cognitive indices, up to 60 minutes
What is freeEverything — score, percentile, four-domain breakdown, full write-upA short free version; the full assessment is paid
What is paidNothingListed at $50 for the basic and $150 for the full test, frequently discounted
Norms and methodA published methodology page describing the scoring model and its limitsNormed on US-born native English speakers aged 18 and over; states adherence to APA, AERA and NCME standards
Sign-upNot requiredAn account is required for the paid assessments
InterfaceLight, modern, original illustrations, mobile-firstProfessional and academic in tone, with a named developer and an expert review panel
How to read this table

None of these rows measure accuracy, because none of them can. What they measure is how much of the result you are given, how much the site tells you about its own method, and what it costs you — in money, in data, or in attention — to get there.

Test length and format

Length is the single biggest driver of how much a reasoning test can tell you, and it is the one specification that is easy to check before you start. Reliability rises with the number of items; the standard error of measurement falls roughly with the square root of test length. That is not a matter of opinion about which site is nicer, it is arithmetic.

riotiq.com runs 15 subtests across six cognitive indices, up to 60 minutes. iq-test.io runs thirty-three questions with no time limit, which puts both of them in the range where a broad band is defensible and a precise number is not.

The timer question matters more than people expect. A test that runs on a clock is partly measuring processing speed, which is a real cognitive ability but a different one from reasoning. If you want to know how well you reason when you are not being rushed, an untimed test answers that question and a timed one answers a blended version of it. Neither approach is wrong; they simply produce scores that are not interchangeable.

What you get without paying

This is where the two sites differ most sharply, and it is worth being precise about it rather than trading slogans.

On RIOT IQ, a short free version; the full assessment is paid comes free. Listed at $50 for the basic and $150 for the full test, frequently discounted is the paid layer.

On iq-test.io, the score, the percentile, the breakdown across four cognitive domains and the accompanying explanation are all free, and there is no paid layer at all. That is unusual enough in this niche to be worth saying plainly: there is nothing to buy, so there is no incentive to hold anything back to make buying attractive.

The pattern to watch for on any IQ site is the gap between the word free in the title tag and the point at which the result appears. A test is free when the score is free. A test that is free to take and paid to see is a paid test with a long preamble.

The certificate question

Several sites in this comparison sell a certificate with your name and score on it. It is worth being clear about what that document is: a decorative PDF generated by the same site that ran the test. No institution, employer or high-IQ society accepts one. If you like having it, that is a perfectly good reason to buy it. There is no other reason.

Norms, scoring and honesty about both

An IQ score is a rank against a comparison group. Without knowing the group, the number means nothing, which makes the norms section the most informative part of any IQ site and the one most likely to be missing.

RIOT IQ: normed on us-born native english speakers aged 18 and over; states adherence to apa, aera and ncme standards.

iq-test.io publishes a methodology page covering the scoring model, the assumptions behind the score conversion and the limits of an unsupervised administration. It does not claim a representative national sample, because a web test does not have one.

Three things are worth checking on any site in this category, including ours:

  • Does it name a norm group at all? Silence here is the loudest signal available.
  • Does it report an error band? A site that gives you a single number with no interval is either unaware of measurement error or has decided you would rather not know.
  • Does it distinguish its own takers from the general population? Percentiles calculated against people who chose to take an online IQ test are not percentiles against everyone.

Our own position on this is on the methodology page, and the short version is that a ten-item test earns a band and nothing finer.

The experience on a phone

Most people take these tests on a phone, which makes the mobile experience a practical concern rather than a design footnote. Matrix items are the hard case: they have to stay legible at 360 pixels wide without the answer options scrolling off the bottom of the screen.

RIOT IQ: professional and academic in tone, with a named developer and an expert review panel.

iq-test.io renders its figures as vector graphics that scale cleanly and keeps answer targets large enough to hit with a thumb. In a test where a misread figure costs you an item, that is a scoring issue and not a cosmetic one.

Who each one is for

Pick iq-test.io if you want the whole result for nothing, you would rather not hand over an email address, and you want a site that documents how it arrived at your score.

Pick RIOT IQ if you want the closest thing to a real psychometric instrument without booking a psychologist.

Pick neither if you need a result that carries weight somewhere. For a school placement, a workplace accommodation or anything legal, an individually administered test such as the WAIS or the WISC is the only route, and no online test substitutes for it at any price.

Reviewed by the IQTest EN editorial team Every reasoning item on this site is solved independently by two reviewers before it is published, and every number in a reference table is traced back to a named source. Where the research is contested, we say so on the page instead of picking the tidier answer. How we write and check this.

Questions

Questions people ask about this comparison

Is RIOT IQ free?

Partly. A short free version; the full assessment is paid is available without paying. Listed at $50 for the basic and $150 for the full test, frequently discounted sits behind a payment. By contrast iq-test.io does not charge for any part of the result.

Is RIOT IQ accurate?

No unsupervised online test is accurate in the sense people mean when they ask. Normed on US-born native English speakers aged 18 and over; states adherence to APA, AERA and NCME standards. That is more or less the ceiling for a web test, and it is why we treat every result on this page as a band rather than a number. The longer answer.

Which is better, RIOT IQ or iq-test.io?

For a free result with nothing withheld, iq-test.io. RIOT IQ is the better choice if you want the closest thing to a real psychometric instrument without booking a psychologist.

Will either result get me into Mensa?

Neither. Mensa uses its own supervised test or a prior result from a short list of approved instruments. How Mensa entry actually works.

Can I compare my scores across the two tests?

Only loosely. Different item banks, different lengths and different norm groups mean a few points of difference between two sittings carries no information at all. If the two results land in the same band, that is agreement; if they land three bands apart, one of the tests has a norms problem.

See what a full free result looks like

Thirty-three questions, no timer, nothing held back, and a methodology page that explains the scoring.