Random Name Generator
Realistic first and last names, filtered by culture and gender, seeded so the same list can be reproduced.
Same seed and settings always give the same list.
Everything here runs on your device. Nothing you enter is uploaded or stored.
Generate realistic first-and-last-name combinations filtered by culture and gender, in batches up to 200, with a seed so the exact same list can be reproduced later. Built for test data, tabletop games, placeholder content, and anywhere else you need names that are not your own.
How to use it
- Choose a culture: English, Spanish, Japanese, French or Indian.
- Filter by gender, or leave it on Any for a mixed list.
- Set how many names you want, up to 200, and pick a seed number.
- Press Generate, then Copy list to grab the whole batch.
- Press Randomise seed for a fresh, unrelated list without changing anything else.
Why the seed matters more than it looks
Most name generators throw away their randomness the moment you refresh the page — ask for a list twice and you get two different lists, with no way back to the first one. That is fine for a single use, but it breaks down the moment you need to share a list with someone else, regenerate a dataset for a second test run, or simply liked the third name in a list you already closed.
This tool uses mulberry32, a small, fully specified pseudo-random number generator. Given the same seed, the same culture, the same gender filter and the same count, it produces the identical sequence of names every time — on any device, any browser, any day. Sharing “seed 4821, English, any, 20 names” with a colleague hands them the exact list, not an approximation.
How the culture lists are built
Each culture’s name pools are hand-curated from that culture’s most commonly used given names and surnames, not scraped from a phonebook or padded with exotic outliers. English draws from decades of common Anglophone first and last names; Japanese pairs common modern given names with the most frequent surnames in Japan; and so on for Spanish, French and Indian. The lists are deliberately real names rather than syllable-generated approximations, because a name that sounds “sort of Japanese” but isn’t a real name serves nobody — not a game designer, not a QA tester, not a novelist naming a minor character.
What it does not do
It does not verify that a generated “First Last” combination is not, coincidentally, a real living person’s name — with common first and last names, some overlap with real people is statistically inevitable, the same as it would be picking names from a phone book. It has a fixed set of five cultures rather than dozens, because each one is curated rather than auto-generated. It does not weight names by popularity within a culture, so a very common name and a less common one from the same pool are equally likely to appear.
Typical uses for a seeded name list
| Use case | Why seeding helps |
|---|---|
| QA test data | Reproduce the exact same dataset across test runs |
| Tabletop RPG NPC roster | Regenerate the same roster if a session is replayed |
| Mockup and prototype content | Share the seed instead of pasting 50 names in Slack |
| Classroom name-draw exercises | Every student sees the same list from the same seed |
Culture coverage
| Culture | Given names | Surnames |
|---|---|---|
| English | 36 | 20 |
| Spanish | 30 | 20 |
| Japanese | 24 | 20 |
| French | 24 | 20 |
| Indian | 24 | 20 |
Questions
What does "seeded" mean here?
The generator uses a deterministic pseudo-random number generator (mulberry32), so a given seed number always produces the exact same sequence of names for the same culture and gender settings. Change the seed and you get a different list; keep it the same and the list is reproducible days or weeks later.
Why would I want a reproducible list?
If you hand a name list to someone else — a test dataset, a tabletop game roster, a placeholder set for a mockup — sharing the seed and settings lets them regenerate the identical list rather than you having to paste the text.
Are these real people's names?
The names are drawn from common real first and last names for each culture, not from any individual's data. Combinations are generated by pairing a first name with a surname independently, so any specific "First Last" pairing is coincidental rather than sourced from a real person.
Why only five cultures?
Each culture's name pool is curated by hand rather than scraped, to keep the names genuinely representative rather than a stereotype list. More cultures are planned; five is the honest starting set rather than a padded list of thin data.
Last updated