Turkish name gender
Turkish names resolve accurately, carry no counted evidence, and include a well-defined set of genuinely unisex names. All three facts matter for how you use the result.
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.
Correct, and unverified
Turkey publishes no name registration counts, so every Turkish answer returns total_names: 0 and confidence: unverified. That is a statement about auditability, not about correctness — the measured accuracy above is the number that speaks to whether the answer is right.
If your filter is "reject anything unverified", you will reject Turkey entirely along with most of the world outside seven Western countries. Filter on probability and on the measured accuracy for the market instead.
Real output, country hint TR
Deniz at 75 is the honest answer for a name used across genders in Turkey. Names in this class — Deniz, Özgür, Evren, Umut, Yücel — are where a country-blind, threshold-free pipeline quietly produces wrong salutations.
| Input | Gender | Probability |
|---|---|---|
| Ayşe | female | 95 |
| Mehmet | male | 95 |
| Zeynep | female | 95 |
| Yağmur | female | 95 |
| Can | male | 95 |
| Deniz | female | 75 |
Dotted and dotless i, and the rest of the alphabet
Turkish distinguishes i/İ from ı/I, and a naive lowercase turns İ into a character that no longer matches. Accent folding here is script-aware rather than a lookup table, so Ayşe, AYŞE and ayse reach the same entry, and ç, ğ, ö, ş and ü behave the same way.
Your input is echoed back unchanged in q, so folding never rewrites the value you store.
Compound and second names
Two-part given names are common — Ayşe Nur, Mehmet Ali, Fatma Zehra. The first token is classified, which is the right choice for these pairs since the first element carries the gender. Where a record holds the full name including the surname, check the name field to confirm the extraction landed on a given name.
One caveat on the measurement at the top of this page: it rests on 61 Turkish names in the fixture. That is enough to show the market works and not enough to quote to three decimal places. If Turkish names are a large share of your data, run your own sample before relying on the figure.
For the unisex class specifically, a probability threshold does most of the work. Accepting only answers at 90 or above removes Deniz and its relatives from your salutation logic while leaving the ordinary names untouched, which is usually the right trade for a mailing and the wrong one for a headcount report.
Questions
Is the site available in Turkish?
The product interface and documentation are, in Turkish and twenty other languages. These landing pages and the notes are English only.
Why unverified when the answer is right?
Because Turkey publishes no registration counts. Confidence measures counted evidence; accuracy is measured separately and published above.
How are unisex Turkish names handled?
They return a probability in the 70s rather than a forced pick. Set a threshold and route the rest to review.
Do Turkish characters need encoding?
Send UTF-8 and percent-encode in query strings. Case and accent folding happen on our side.
Related pages
Predict gender from an Arabic name written in Arabic script or transliterated. How the consonant skeleton index works and why it beats guessing a spelling.
Predict gender from an Indian name in romanised or Devanagari form. This is our strongest non-Latin market, and this page shows the measured numbers.
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.
Some ordinary names flip gender across borders, and the error is systematic rather than random. What the country parameter does, and where the data comes from.
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.