Highest IQ ever recorded
On this page
- The short answer: there is no official record
- Ratio IQ vs deviation IQ: why old scores look so huge
- Marilyn vos Savant and the 228 that Guinness retired
- Other famous names, and how solid their numbers are
- Why standard tests stop around 160
- The practical benchmark: the top 2%
- Frequently asked questions
Ask who has the highest IQ in the world and the internet answers confidently: 228, 250, even 276. Here is the awkward truth behind those numbers: there is no official record, the organisation that once kept one gave it up, and the most famous scores of all were produced by a method modern testing abandoned decades ago. This page explains where the giant numbers come from, which names are attached to them, and where reliable measurement actually stops in practice.

The short answer: there is no official record
Guinness World Records used to list a "Highest IQ" category. It retired the category in 1990, concluding that IQ tests were too unreliable to designate a single record holder. That decision, from the people whose business is records, is the most honest answer to "what is the highest IQ?" available: beyond a certain point, the instruments cannot crown a winner. You can read more about how records and categories work at Guinness World Records.

Everything else, every list of "top 10 highest IQs ever", is assembled from old scores, estimates, and claims of very different quality. To see why, you need one key distinction.
Ratio IQ vs deviation IQ: why old scores look so huge
Early IQ tests used ratio IQ: mental age divided by chronological age, multiplied by 100. A child of 10 performing like an average 15-year-old scored 150. The catch is that this formula explodes for bright children. Small differences in "mental age" turn into enormous numbers, and it makes no sense for adults at all.
Modern tests use deviation IQ instead. Your performance is compared with a large sample of people your age, the average is set to 100, and 15 points equal one standard deviation. This is the system behind the scores in our IQ score chart and behind the fact that the average IQ is 100 by definition.
The two systems produce numbers that look alike but are not interchangeable. Nearly every three-digit headline score above 200 is a ratio score, an extrapolation, or a score from an unofficial test, not a deviation IQ from a standardised professional test.
Marilyn vos Savant and the 228 that Guinness retired
The most famous case is Marilyn vos Savant, the American writer known for the "Ask Marilyn" column. As a child in 1956 she took an old Stanford–Binet test and was credited with a mental age of 22 years and 10 months at age 10, which, by the ratio formula, gives 228. Guinness listed her under Highest IQ from 1985 to 1989, then retired the category in 1990.
None of this means vos Savant is not exceptionally intelligent, she plainly is. It means the number 228 belongs to an abandoned scoring method applied to a child, and test experts have long pointed out that the old test's own norms did not support scores anywhere near that high. Her 228 is a historical artifact, not a score that can be compared with a modern test result.
Other famous names, and how solid their numbers are
William James Sidis (claimed 250–300). The American child prodigy of the early 1900s is often credited with an IQ between 250 and 300. There is no reliable test record behind these figures; they are estimates repeated so often they hardened into "fact." Treat them as legend, not measurement.
Terence Tao (estimated ~225–230). The UCLA mathematician, a Fields Medal winner and one of the most respected living mathematicians, is frequently said to have an IQ around 225 to 230. Tao did score exceptionally on ability tests as a child, but the widely circulated figures are estimates, not a published, standardised adult score. His UCLA faculty page documents the achievements, which need no IQ number to be impressive.
Kim Young-hoon (claimed 276). In recent years the South Korean has been reported with a score of 276 from a non-standard high-range test. The claim is disputed: the test involved is not a professionally standardised instrument, and the figure has not been accepted by Guinness or the wider testing community. A deviation IQ of 276 would imply a rarity far beyond the number of humans who have ever lived, which is itself a reason for scepticism. Reporting by Vice adds telling detail: the 276 figure was produced by converting a score of 210 on the non-standard VNPT-II test onto a standard deviation 24 scale, and that test was normed on a sample of only 32 people. Experts have disputed the claim.
Christopher Hirata and others (~225 claims). Several other names circulate with scores in the 190s–220s, typically from high-range hobbyist tests taken as teenagers or adults. These tests are designed by enthusiasts, taken by small self-selected groups, and normed on tiny samples. Scores from them cannot be compared with professional results.
Why standard tests stop around 160
On a deviation scale, each step up is rarer than the last. A score of 130 is about top 2%. A score of 145 is about 1 in 740. A score of 160 corresponds to about 1 in 30,000 people. To norm that level properly, to know exactly where it sits, a test publisher would need standardisation samples of hundreds of thousands, which no professional test has. In practice, instruments like the WAIS cannot reliably distinguish scores much above 155–160. Claims beyond that are not measurements; they are extrapolations. You can look up the rarity of any score in our IQ percentile calculator. This is also why questions about extreme scores belong with questions about how much any test can really tell you, see are online IQ tests accurate?.
The practical benchmark: the top 2%
If "highest IQ in the world" has no clean answer, what benchmark actually means something? The one used by high-IQ societies: the top 2% of the population, which works out at about 131 on a test with a standard deviation of 15. That is the threshold Mensa International uses, not a single world-beating number, but a percentile anyone can in principle verify with a properly administered test. We explain the details in our guide to the Mensa IQ requirement. It is a less exciting answer than 276, but unlike 276, it is a real one.
Frequently asked questions
What is the highest IQ possible?
There is no theoretical maximum written into the scale, but there is a practical measurement ceiling. Standard professional tests run out of reliable norming at around 160, so any number far above that is not a verified measurement, whatever its source.
Who has the highest IQ in the world today?
Nobody knows, and there is no body that certifies it. Guinness retired the category in 1990. The people named in popular lists hold old ratio scores, estimates, or disputed claims from unofficial tests, not comparable, verified records.
Is an IQ of 200 possible?
On paper, a deviation IQ of 200 would describe someone rarer than about 1 in 76 billion, essentially unmeasurable with real norming samples. No standardised professional test has ever verifiably produced such a score.
What IQ do I need to be in the top 2%?
About 131 or higher on a test with a standard deviation of 15 (the exact cut-off depends on the test Mensa accepts). That places you above 98% of the population, a genuinely rare level, even if it sounds modest next to the mythical 200s.
Curious about your estimated score? Try our free IQ test, with 15 questions in 10 minutes or 40 in 30 minutes. Get an estimate ranging from 55 to 145 score along with explanations for every answer. No account or email required.
See where you land on the scale.
Take the free test, 15 puzzles in 10 minutes or 40 in 30, and get your estimated IQ with every answer explained, plus a free certificate with your name. No sign-up, nothing to pay.