Name-to-gender credits for academic research
If you study authorship, citation gaps, grant outcomes or any other question that needs gender inferred from names, we give 100,000 free credits to your project. In return, we ask you to cite NameGender in what you publish.
100 free credits every day. No card.
What you get
One grant of 100,000 credits on your account. One credit is one name, through the API, the bulk endpoint or a spreadsheet upload in the dashboard. Granted credits stay on your balance until you spend them, so a study that runs over several semesters does not lose them at the end of a month.
This sits on top of the free tier every account already has: 500 credits on signup and 100 a day after that, which is enough to pilot your pipeline before you apply. If your study needs more than 100,000 names, say so in the application and explain the corpus; larger grants are decided case by case.
Why the output suits a methods section
Every answer returns the share of records behind it, the number of records (`sample_size`), a confidence tier derived mechanically from that count, and the source layer it came from. You can report your threshold, the share of names left unknown and the evidence behind each classification, instead of a single accuracy figure from a vendor page.
Each response also carries a `data_version`, so you can state which snapshot of the data produced your results and a reviewer can ask for the same one. Our accuracy measurements, including the per-country and per-script results where we do badly, are published with their fixtures on the benchmark and methodology pages.
Counted data comes from official name statistics in eleven countries. For the rest of the world the evidence is a list of names attested with a gender, without counts, and those answers say so. If your corpus is heavy in Thai, Hebrew or Arabic-script names, read our published accuracy for those scripts before you commit: it is lower than for Latin, Chinese or Korean names.
How to apply
Create a free account with your institutional email address, then open a support ticket with the topic "Academic research programme". The button on this page opens that form with the topic chosen and a short template to fill in: your institution, the project, what the names are used for, roughly how many names, and where you expect to publish.
We read each application ourselves. Doctoral students are welcome; please add your supervisor's name. The programme is for non-commercial research. Work done for a company, or a study whose main output is a product, is outside it.
What we ask in return
Cite NameGender wherever the inferred genders feed into a published result: “NameGender (2026). Name-to-gender API, data version 2026.10. https://namegender.com”. Replace the data version with the one your responses returned. If you also use our open dataset of names that change gender by country, cite it by its DOI, 10.5281/zenodo.23126656.
When the paper, thesis or preprint is out, send us the reference in the same ticket. Nothing else is required: no logo, no review, no endorsement. You are free to report results that are unflattering to us, including comparisons with other services.
Handling names responsibly
Author lists and survey records can be personal data. You can create an API key with the setting "Don't store the names this key sends", and the names you send are then never written to our request history; the lookup and your balance work the same way.
A name-based estimate describes how a name has been used in a population, not how any person identifies. It is a reasonable instrument for aggregate questions, such as the share of women among first authors in a field. It is not a basis for decisions about individuals, and we will not grant credits to projects that use it that way.
Questions
How many free credits do researchers get?
One grant of 100,000 credits per project, on top of the free tier of 500 credits on signup and 100 a day. Granted credits do not expire. Larger studies can ask for more in the application.
Who can apply?
Researchers and doctoral students at universities and non-profit research institutes, for non-commercial work. Apply from an account registered with your institutional email address.
How should I cite NameGender?
NameGender (2026). Name-to-gender API, data version 2026.10. https://namegender.com — with the data version your responses returned. The open dataset of names that change gender by country has its own DOI: 10.5281/zenodo.23126656.
Can I report results that are bad for NameGender?
Yes. The grant does not depend on what you find, and we do not review your paper before publication. We publish our own weaker results, such as Thai and Hebrew names, on the benchmark page.
Related pages
What is in the name-to-gender dataset: counted registries from eleven countries, WGND 2.0 for the rest, versioned snapshots and a per-answer source field.
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.
Upload a CSV or XLSX name list and get a gender column back, with confidence fields, automatic deduplication and the cost shown before charging.
Measured on 4,610 labelled names from 25 countries: 94% answered, 97% of answers correct, 91% end to end. Results by script, and where the errors come from.
Reading the gender field alone discards everything that says whether to believe it. What each response field means and the thresholds behind confidence.
Turn on one setting per API key and the names you send are never written to our request history. What is still recorded, and why a hashed name is not anonymous.
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.