More thanjust a name.
Send a name or an email address. Get the gender with its probability and the sample size behind it, the first and last name split apart, and a greeting ready to paste: “Dear Ms. Johnson:”, “Sehr geehrter Herr Weber,”, “Estimada Sra. García:”.
Names that change by country
Male
Female
Female
Female
Female
Country-by-country figures for any name come from the /countries endpoint.
Paste up to 100 names or email addresses, or open a CSV file. You get gender, first and last name and a ready salutation for each row. No signup. The file stays in your browser; only the column you pick is sent.
Only the first 100 rows are used here. A free account processes the whole file.
| input | first_name | last_name | gender | probability | salutation |
|---|---|---|---|---|---|
Below 90% probability the salutation stays neutral instead of guessing.
A free account processes 100 names a day, no card required, and the file upload keeps every column you already have.
The labels come from public name lists we did not compile, and the test runs the same code as the API. We publish the writing systems we handle badly next to the ones we handle well.
In the NameGender 2026 benchmark of 4,610 labelled names from 25 countries, the API answered 94% of the names and 97% of those answers were correct, so 91% of all names were answered correctly.
Share of all test names in each writing system answered correctly. Unanswered names count as misses.
98% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
98% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
98% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
100% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
100% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
80% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
66% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
51% answered · Share of all test names in each writing system answered correctly. Unanswered names count as misses.
100% answered
100% answered
100% answered
99% answered
100% answered
98% answered
100% answered
98% answered
100% answered
100% answered
80% answered
66% answered
98% answered
96% answered
100% answered
100% answered
100% answered
95% answered
100% answered
98% answered
100% answered
51% answered
98% answered
100% answered
89% answered
One API, four input types, and bulk tools that handle files other services choke on.
One REST API for names, emails and usernames, the same JSON shape for all three, and a bulk endpoint for up to 100 values per request.
curl "https://namegender.com/api/v1/gender?name=Jean&country=FR" \
-H "Authorization: Bearer $NAMEGENDER_API_KEY"
{
"name": "Jean",
"gender": "male",
"probability": 100,
"sample_size": 1910805,
"country": "FR",
"confidence": "high",
"source": "db"
}Run a spreadsheet job: send an XLSX or CSV file, tell us which column holds the names, and your rows come back with gender, probability and sample size alongside. No code.
| first_name | country | gender | probability | based_on |
|---|---|---|---|---|
| Jean | FR | male | 100 | 1910805 |
| Robin | US | female | 86 | 338093 |
| Andrea | US | female | 99 | 450034 |
| Kim | US | female | 84 | 214713 |
Compound names and names with titles or initials are cleaned up before the lookup, so you can send the field as it is.
ayse.yilmaz84@example.com is read as Ayşe; role mailboxes such as info@ come back unknown.
camelCase, snake_case and trailing digits are split apart before the lookup.
Give us an XLSX or CSV file and get the same rows back with gender columns added. 200,000 rows take under a second of compute.
Andrea is male in Italy and female in Germany. Pass a country and we weigh it.
Every answer tells you where it came from and how large the sample was.
Buy 20,000 credits for €12.00, or 10,000,000 for €333.00. Whatever you do not use stays on your account: no monthly reset and no inactivity clause.
Every account also gets 100 free credits a day.
See all packages| Package | Credits | Price | Price per 1,000 names |
|---|---|---|---|
| 20K | 20,000 credits | €12.00 | €0.600 per 1,000 |
| 10M | 10,000,000 credits | €333.00 | €0.033 per 1,000 |
A gender API predicts the gender most often associated with a first name, using counted birth registrations and other name statistics. NameGender returns the gender together with the probability, the number of people the answer is based on, the country and the source, and it returns unknown when the evidence is too thin.
In the 2026 benchmark of 4,610 labelled names from 25 countries, NameGender answered 94% of the names and 97% of those answers were correct, so 91% of all names were answered correctly. Names missing from the dataset altogether are a different case: of 2,000 names removed from it for a separate test, 10% were answered, by closest spelling, and 97% of those answers were correct.
Yes. Turkish names use the Latin script, where 96% of test names were answered correctly. Arabic, Chinese and Korean names are matched directly or after transliteration: 96%, 91% and 92% of test names were answered correctly. Hebrew and Thai names are answered less often: 64% and 49% of test names were answered correctly.
Yes. Jean is recorded as male in France (100%) and as female in United States (95%). Pass a two-letter country code and the answer uses that country's data; without one you get the worldwide aggregate.
Yes. The email endpoint looks up the first name in the local part, so john.doe@example.com resolves as John, and the username endpoint strips digits and separators first. Role addresses such as admin@ or info@, placeholder handles such as user123, and an initial followed only by a surname, as in j.smith@, come back unknown rather than as a guess.
Purchased credits stay on your balance until you spend them. Packages start at €12.00 for 20,000 credits, the largest costs €0.033 per 1,000 names, and every account gets 100 free credits a day.
No. A name-based result is a statistical association, not a person's gender identity. Use it for aggregate work such as audience segmentation or survey weighting, keep unknown results visible, and do not use it for employment, medical, financial, insurance, legal or eligibility decisions.
Measured coverage and worked examples for each. Written in English only.
No credit card. Anything you buy stays on your balance until you spend it.
Create your free accountWith your permission we measure how the site is used and which adverts lead to a sign-up. Both are optional and neither runs until you allow it. Read the cookie policy.