Korean name gender
Hangul and romanised input both resolve. Coverage is near total; accuracy sits below the European figure for the same structural reason as Chinese — Korean given names are composed rather than drawn from a fixed list.
100 free credits every day. No card.
Measured 28 August 2026 on a fixed, reproducible fixture. Not a population sample — a fixed input anyone can rerun against us or against a competitor.
Composed names, syllable by syllable
A Korean given name is usually two syllable blocks chosen for meaning and sound. Some blocks lean strongly male or female; many are used for both, and the pairing shifts the reading. That is why probability carries real information here and the gender field alone does not.
지우 comes back female at 71 — not a rounding artefact, an accurate description of a name that is genuinely used across genders.
Real output, country hint KR
Hyun at 60 is the honest shape of a single syllable submitted without its partner: barely more than a coin flip, and reported as such.
Korean answers carry total_names: 0 and confidence: unverified because Korea publishes no name registration counts. That is a statement about auditability rather than correctness — the measured accuracy at the top of this page is the figure that speaks to whether the answer is right.
The gap between near-total coverage and a lower accuracy is the same structural fact as in Chinese: we can almost always find the name, and the name itself carries less gender signal than a European one does. Treat probability as load-bearing here rather than reading the gender field alone.
| Input | Gender | Probability |
|---|---|---|
| 지훈 (Ji-hoon) | male | 95 |
| 서연 (Seo-yeon) | female | 95 |
| 민준 (Min-jun) | male | 95 |
| 지우 (Ji-woo) | female | 71 |
| Min-jun | male | 90 |
| Hyun | male | 60 |
Romanisation is not one system
Revised Romanisation and McCune-Reischauer disagree, and personal preference overrides both on passports. The same name appears as Ji-hoon, Jihun, Jihoon and Ji Hun in real datasets. Hangul input avoids the whole problem; when you only have romanised text, expect some of your misses to be spelling variants rather than unknown names, and check matched_as to see when a near-spelling was used.
Surname first, and short surnames
Korean names are written surname first, and the common surnames are single syllables — Kim, Lee, Park, Choi. 박서준 sent whole is matched as a unit; Park Seo-jun sent romanised risks classifying Park. Send the given name alone where you can, and read back the name field where you cannot.
Questions
Does hyphenation matter?
No. Hyphens, spaces and joined forms all normalise to the same lookup.
Why is a correct Korean answer marked unverified?
Korea publishes no name registration counts, so answers carry zero observations. Confidence reports counted evidence, not correctness.
Hangul or romanised — which should I send?
Hangul, when your system stores it. It removes the romanisation variance entirely.
What about North Korean or diaspora naming?
The same lookup applies; diaspora records often carry a Western given name instead, which resolves on the Latin data and usually with counts behind it.
Related pages
Predict gender from a Japanese name. Our measured accuracy here is the lowest of any market we publish, the sample is small, and this page shows both.
Predict gender from a Chinese name in Han characters or pinyin. Measured coverage and accuracy, the surname-first trap, and why characters beat romanisation.
Every name lookup returns a probability, a sample size and a confidence tier. What each one measures, how the tiers are set, and where to put your threshold.
Coverage on Han names is 99.5% but accuracy is 86.9% — the inverse of the usual pattern. Why pinyin is weaker evidence, and the name-order trap.
A reproducible benchmark across 4,610 names, 25 countries and nine writing systems. Two scripts return almost nothing, and we publish those numbers too.
Check it against your own list
Every number on this page is reproducible with a free key. If your data breaks it, that is the more interesting result.