Realistic test names from Faker's English locale — seeded, exportable, in-browser
Quick answer: this generator draws real-world-style names from Faker's English locale — 3,186 first names × 473 surnames = 1,506,978 distinct first-last pairs, roughly 82.9M once prefixes, middles, and suffixes join in. Set a seed (say 2026) and you get the same five names every run: Ms. Vivienne Kris V, Genoveva Walker, Shania Lebsack, Alberto Ebert, Alyce Ziem. No seed, fresh names each click. CSV and JSON export included, and nothing uploads.
| # | Name |
|---|---|
| Loading Faker engine… | |
| Pool | Size | Examples from the data |
|---|---|---|
| Female first names | 1,563 | Vivienne, Genoveva, Shania, Alyce |
| Male first names | 1,391 | Gerardo, Alberto, Damion, Ali |
| Genderless first names | 232 | Kris, Jordan, Taylor class names |
| Surnames | 473 | Beahan, Cormier, Lebsack, Zieme, Powlowski |
| Prefixes | 5 | Dr., Mr., Mrs., Ms., Miss |
| Suffixes | 11 | Jr., Sr., I, II, III, IV, V, MD, DDS, PhD, DVM |
Counts were measured directly from the locale data this page loads (@faker-js/faker 10.6.0, en). Combos = first pool × surname pool; the 82,883,790 full-name space multiplies in 5 prefixes × 11 suffixes. A 10,000-name run produced 9,981 distinct names — about a 0.19% duplicate rate, right where birthday-paradox math puts it.
The engine is @faker-js/faker 10.6.0 (MIT), the open-source test-data library behind thousands of test suites. This page lazy-loads exactly one locale bundle — English — the first time you generate, then draws names locally in your tab. No name you generate is looked up, stored, or sent anywhere.
Faker's draws come from a seeded pseudo-random generator. Type a seed and the sequence starts from a fixed point: identical inputs, identical names, forever. That's why the worked example below never changes, and why seeded fixtures are the standard trick for flaky-test debugging. Leave the seed blank and every click starts from a fresh random state.
Pick a count up to 500, filter by gender if your test cases need it, choose a format (full names include an occasional middle name, prefix, or suffix — that's Faker being realistic, not a bug), and generate. Copy the table as CSV for spreadsheets, as JSON for fixtures files, or download the CSV directly. The gender filter is pool-level: it draws only from the 1,563 female or 1,391 male names, plus the 232 genderless ones.
Type 2026 in the seed box, set count to 5, gender Any, format Full name, and hit Generate. Out comes:
Switch gender to Female with the same seed and the list becomes Ms. Ida Borer-Kris V, Elyse Grimes Jr., Lauren Powlowski Jr. — different draws, same reproducibility (yes, Faker will happily hang a Jr. on a female-mode name; suffixes draw independently of gender, matching real-world data mess). The seed box is your fixture identity: store it next to the test, not the names.
Need the rest of a realistic record to go with the name? Our random address generator and fake email generator run on the same engine and seeds.
Yes, occasionally — with 1,506,978 possible first-last pairs from this generator's pools, some combinations belong to real people, and with millions of people on Earth, most plausible names are shared. That's fine for form testing and demo data. It is not fine for impersonation, synthetic personas meant to pass as real, or anything where a reader might believe the person exists. If a generated name happens to match someone you know, reroll — every click without a seed produces a fresh set.
A seed restarts the generator's random sequence at a fixed point, so the same seed always produces the same names. Seed 2026 with default settings yields Ms. Vivienne Kris V, Genoveva Walker, Shania Lebsack, Alberto Ebert, and Alyce Ziem — today, tomorrow, and on any machine, because the same locale data and generator version are loaded. That reproducibility is what makes seeded fixtures useful in test suites: a failing test can regenerate the exact data that broke it.
The English locale packs 3,186 first names (1,563 female, 1,391 male, 232 genderless) and 473 surnames — 1,506,978 distinct first-last pairs. Full names add a middle name sometimes, plus 5 prefixes and 11 suffixes, for about 82.9 million combinations. In practice we measured 9,981 distinct names in a 10,000-name run: about 0.19% duplicates, which is the birthday-paradox math you'd expect, not a flaw.
This generator loads Faker's English (en) locale, whose name pools lean American and British — Murphy, Kris, Beahan, Cormier. Faker ships 70+ other locales (German, Japanese, French, and more) but each adds download weight, so this page sticks to one. For localization testing, generate with the en locale and add locale-specific entries at the data layer, or self-host Faker with the locales you need.
Synthetic names aren't personal data the way real customer records are, which is exactly why test teams generate them — but privacy law compliance depends on your context, the other fields you pair them with, and your jurisdiction. Pairing a fake name with a real email inbox or real address can re-create a personal-data problem. Treat generated names as one safe ingredient, and check your data-handling rules with whoever owns them; this tool is engineering help, not legal advice.
Yes — every run can be copied or downloaded as CSV (one column per displayed field) or as JSON (array of objects). The CSV quotes fields containing commas, so names like Anne-Marie Smith Jr. paste cleanly into spreadsheets, and the JSON shape matches what most fixtures files expect. Because the export is generated from the same seed shown in the page, your fixtures stay reproducible.