Random Name Generator
Discover unique names with our collection of 27 different generators. Real census data, curated word banks, instant results — completely free.
27 Name Generators
From real surnames to fantasy realms — find the perfect name for any occasion.
Random Name Generator
First and last names, totally random
Generate →Last Name Generator
5,000+ real surnames from census data
Generate →Nickname Generator
Cool nicknames in one click
Generate →Username Generator
Unique usernames for any platform
Generate →Couple Name Generator
Blend two names into one
Generate →Team Name Generator
Names that strike fear (or fun)
Generate →Fantasy Name Generator
Elvish-sounding names for your world
Generate →Town Name Generator
Charming village & city names
Generate →Gamer Tag Generator
Stand out in the lobby
Generate →WiFi Name Generator
Funny names for your router
Generate →Instagram Username Generator
Aesthetic handles for Insta
Generate →TikTok Username Generator
Viral-ready TikTok handles
Generate →Discord Username Generator
Handles for your server
Generate →Baby Boy Name Generator
Popular & classic boy names
Generate →Baby Girl Name Generator
Beautiful girl names
Generate →Elf Name Generator
Elegant elven names
Generate →Dwarf Name Generator
Sturdy dwarven names
Generate →Pirate Name Generator
Yarr! Pirate aliases
Generate →Medieval Name Generator
Names from the Middle Ages
Generate →Cowboy Name Generator
Wild West aliases
Generate →Anime Name Generator
Japanese-style character names
Generate →K-Pop Name Generator
Your stage name awaits
Generate →Rapper Name Generator
Your hip-hop alias
Generate →Band Name Generator
From garage to stadium
Generate →Superhero Name Generator
Cape-ready hero names
Generate →Villain Name Generator
Delightfully evil names
Generate →Japanese Last Name Generator
Common Japanese surnames
Generate →Explore Name Lists
Hand-curated collections, ready to browse.
Cool Surnames
200+ effortlessly cool surnames
Unique Last Names
Rare gems you won't hear twice
Victorian Surnames
Step back into the 1800s
Japanese Last Names
60 most common surnames
French Last Names
Classic French surnames
Nordic Last Names
Scandinavian heritage names
Fantasy Names — Male
Heroic names for your champions
Fantasy Names — Female
Enchanting names for heroines
Name Tools
Pick a winner or shuffle a list — visual and fair.
Why "random" is the whole point
Ask someone to invent a name under pressure and watch what happens: they reach for whatever floated past them last week. Novelists default to Jake and Sarah. Developers seed test databases with John Doe a hundred times in a row. Game masters improvise "Bob the innkeeper" and then spend the rest of the campaign apologizing for it. The failure is not a lack of imagination — it is that naming from memory collapses into a tiny, overused sample of names, and our brains pull from that same sample every single time.
A random name generator fixes the input problem instead of the imagination problem. It hands you twelve options you did not think of, and your judgment — which is actually good at rejecting bad names — does the rest. That is the loop this site is built around: generate a batch, keep one or two, reroll the rest. It takes about ten seconds per round, and it beats staring at a blank character sheet by a wide margin. There is a real academic field behind all of this, too — onomastics, the study of names — and a few of its findings are genuinely useful, so we will sprinkle them in below.
What the 2024 baby name data actually says
The US Social Security Administration publishes annual baby name statistics from its nationwide registrations — well over three million newborns a year — and the 2024 release is a good window into how naming actually behaves. The top ten boys' names: Liam, Noah, Oliver, Theodore, James, Henry, Mateo, Elijah, William, and John. The top ten girls' names: Olivia, Emma, Charlotte, Amelia, Sophia, Mia, Isabella, Ava, Evelyn, and Luna. Two things stand out. John — the most-issued boys' name for most of the twentieth century — clawed its way back into the top ten for the first time in decades, and Luna rode the momentum of a generation of children's books and one very popular Beatles song into the girls' list.
Here is the number that matters more than any top-ten list: even the #1 name, Liam, goes to only about 1 in 100 baby boys. Olivia lands near 1 in 110 girls. The classic image of a classroom full of four Jennifers — a genuine 1980s phenomenon, when Jennifer peaked at over 4% of all girls — simply does not exist anymore. Name choice has fragmented hard since the 1990s, which means the statistically "normal" name today can be almost anything. Our baby boy and baby girl generators lean on that reality: a spread of familiar-but-not-top-ten names, because that is what an actual 2025 kindergarten roster looks like.
The 1% surname: what census data teaches
Surnames obey a harsher math than first names. According to the US census, about 2.4 million people — very nearly 1% of the country — are named Smith. Widen the lens and it gets stranger: the ten biggest surnames (Smith, Johnson, Williams, Brown, Jones, Garcia, Miller, Davis, Rodriguez, Martinez) cover roughly 1 in 25 Americans, and the top 1,000 surnames cover nearly half the population. Meanwhile tens of thousands of surnames are held by fewer than 100 people each. That lopsided curve — a handful of giants and a long tail of rarities — is called a power-law distribution, and it is why frequency-weighted generation feels different from a plain alphabetical list.
The last name generator samples from exactly that curve. Roll twelve surnames and you will usually see one or two giants and a scattering of mid-frequency names, with a rare one appearing once in a while — which is what a real phone book looks like. If you specifically want the tail, filter by a rare initial: only a few percent of Americans carry surnames starting with Q, X, or Z. Writers use this deliberately. Give the hero's best friend a Smith, give the mystery character a Vandermeer, and readers feel the difference without knowing why.
How similar-name clustering works (KNN, minus the math)
One quiet trick behind modern name tools is a nearest-neighbor search — the same KNN idea that powers recommendation engines. Represent every name as a point based on its features: length, syllable count, which consonants and vowels it uses, where the stress falls, how it ends. Then, given one name you like, find the closest points in the pool. Type in "Isabella" and the neighbors are Arabella, Gabriella, Maribella — not because a database says so, but because those names share the four-syllable shape, the soft consonants, and the -ella ending.
You can run a manual version of this right now without any code. Generate a batch of names, pick the single one that clicks, and then generate several more batches looking only for names that share its skeleton — same ending, same rhythm, same first consonant. Three rounds of that and you have a matched family of five to eight names that sound like they belong to the same culture, dynasty, or fantasy realm. Novelists use this to name whole noble houses; game designers use it to name an entire elf village that sounds internally consistent instead of like twelve unrelated ideas.
Stuck naming a character? Generate, then subtract
Writers hit a specific wall: the protagonist's name has been "TBD" for six weeks and the draft cannot move without it. The productive move is counterintuitive — stop trying to find the right name and start eliminating wrong ones. Generate twenty names, cross off fifteen immediately, and study the five survivors. Almost always the reaction is not "this one!" but "closer, but the ending is wrong," and that reaction is information. Wrong-ending, wrong-era, wrong-country — each crossed-out name narrows the target.
A few field-tested tricks from working novelists. Read the finalists aloud inside a line of dialogue — some perfectly good names turn to mud when a character has to say them three times a chapter. Check the initials: a hero named Brandon Arthur Davis is fine until the monogram reads B.A.D. Vary the syllable count across your cast, because a reader skimming a page will confuse two four-syllable names far more often than a two-syllable and a five-syllable one. And if the book spans generations, remember names age: a woman born in 1935 is a Barbara or a Patricia, not a Brooklyn or a Nevaeh — both of which are top-200 names that simply did not exist in that decade's data.
D&D and game NPCs: naming fifty people before dinner
A single D&D session can chew through a dozen named characters. The party talks to one guard, robs one merchant, angers one mayor, and by the third hour somebody asks the innkeeper's name — and whatever you blurt out becomes permanent canon the players will reference for two years. Experienced game masters handle this with a prepared batch: twelve first names and twelve surnames printed at the bottom of the session notes, so any improvisational NPC gets assembled in two seconds and sounds like a person rather than a placeholder.
Video game teams run the same pattern at industrial scale. A procedurally populated city needs 5,000 residents, which means 5,000 plausible name combinations, deduplicated, filtered for profanity strings (including borderline ones that only turn offensive when a first name sits next to them — every live game learns this once), and weighted so the city's demographics look natural. Indie developers routinely generate the list here, export it as plain text, and paste it straight into a spreadsheet or a JSON array. For tabletop specifically, the name picker wheel doubles as a session tool — load your NPC shortlist and let the wheel decide who the players actually talk to first.
Names across languages: diacritics and the José problem
Try putting José, Zoë, or Müller into a username field built in 2011 and watch it fail. Accented characters — diacritics, in the technical sense — are where international naming falls apart in practice. The letters exist in Unicode and have for decades, but databases, game engines, payment forms, and legacy systems still mangle them. The common workarounds tell their own stories: airlines strip diacritics from boarding passes (José becomes JOSE, Müller becomes MUELLER), databases transliterate non-Latin scripts entirely, and anyone named Zoë has at least one story about an ID mismatch.
For practical purposes, know which world you are naming into. Fiction and genealogy should keep the accents — José without the accent is a small but real signal of carelessness to readers who carry the name themselves. Usernames, API identifiers, and game handles usually want the ASCII-stripped form, which is why most generators on this site output plain letters by default. If you need the accented original, browse the curated pages like French last names or Japanese last names, where original spellings are preserved rather than flattened. And when a name must survive a signup form, the username generator works from ASCII-safe pieces from the start.
The rise of gender-neutral names, 2015 to 2025
If you compare SSA data from 2015 against 2024, one trend is impossible to miss: names shared across genders have roughly doubled their share of births over the decade. Riley, Rowan, River, Quinn, Sage, Kai, and Charlie now appear in meaningful numbers on both the boys' and girls' charts, with Riley alone given to tens of thousands of children of each sex annually. Some names have flipped sides entirely over longer arcs — Leslie and Ashley both started as boys' names a century ago — while newer arrivals like Sage and River were unisex from the start.
The numbers carry real consequences for fiction and design. A 2020s character named Riley tells the reader nothing about gender — which is sometimes the author's exact intent, and sometimes a surprise in chapter four. Casting directors and game designers have noticed the same shift: unisex names are now the default recommendation for placeholder characters precisely because they fit any casting decision made later. If you want that flexibility, generate a pool and keep only the names that do not trigger a gender assumption on first read — with 2020s data, that shortlist fills up faster than it would have with 1980s data, when gendered names were far more rigidly sorted.
A two-minute onomastics starter
Onomastics sounds forbidding but starts with a simple observation: every name is a fossil. English surnames split into four ancient strata — patronymics (Johnson: John's son), trades (Smith, Baker, Fletcher), places (Hill, Wood, York), and nicknames (Fox, Little, Armstrong). Read a surname that way and it stops being a label and becomes a sentence about an ancestor: "the son of John who baked bread near the hill." First names carry their own strata — biblical names, royal names, saint's-day names, and the modern layer of invented spellings that spell-checkers refuse to learn.
Two findings from the field are worth carrying into any naming project. Names ride visible waves: after a hit TV character, usage can multiply tenfold within three years (the name Arya went from a few dozen births per year in the US to over 2,500 within five years of Game of Thrones starting). And names signal generation better than almost any other word choice — a fictional résumé that says "Linda, born 1955" and one that says "Madison, born 1955" contain an immediate contradiction that careful readers will feel before they can explain it. That is exactly the kind of detail a generator with era-aware pools helps you get right.
Usernames and the squished-name problem
A quiet majority of internet usernames are just crushed names: first-initial-plus-surname, surname-plus-birth-year, name-plus-favorite -number. That works until it does not — there are only so many jsmiths and jsmith87s, and the good ones were claimed twenty years ago. The pattern that holds up is contrast: a name plus a word from a completely different domain (fernsushi, harborwolf) reads as intentional, while name-plus-random-digits reads as a default email address. Given that roughly 4 in 10 people report reusing passwords across accounts, a memorable-but-not-guessable handle is worth thirty seconds of thought.
The username generator handles the collision problem directly, and the niche variants go further — gamer tags built for voice chat (short, shoutable, no ambiguous spellings), Discord and Instagram flavors matched to how those platforms display handles. Whatever route you take, run the same two checks as fiction: say it aloud once, and make sure it cannot be misread — the classic lowercase failure being any name containing "lick," "lass," or an unfortunate vowel run. Ask the "Pen Island" domain what happens when you skip that check.
Why RandomNames.top?
We believe finding a good name shouldn't take an afternoon. All of our generators run entirely in your browser with real census data and hand-curated word banks — no sign-up, no limits, no nonsense.
Random Name Generator — FAQ
+Is the random name generator free to use?
Yes. All 27 generators, curated name lists, and the spin-the-wheel picker are completely free, with no sign-up and no usage limits. Everything runs in your browser.
+Where do the random last names come from?
The surname generator draws from real census data, so the last names you get reflect how common names actually are, letter by letter. Other generators use hand-curated word banks built for each category.
+Can I use the generated names for games, stories, or business?
Yes. Names generated here are free to use anywhere — character names for novels and games, usernames, baby name brainstorming, or placeholder data. Real census surnames are public data.
+How does the spin-the-wheel name picker work?
Paste your list of names into the wheel tool, spin it, and it picks one winner at random with a fair shuffle. It is handy for classrooms, raffles, and team assignments.
+What are the most popular baby names in the US right now?
According to the Social Security Administration's 2024 data, the top boys' names are Liam, Noah, Oliver, Theodore, James, Henry, Mateo, Elijah, William, and John. The top girls' names are Olivia, Emma, Charlotte, Amelia, Sophia, Mia, Isabella, Ava, Evelyn, and Luna. Our baby name generators mix popular picks with less common ones so you are not stuck choosing from the same ten names as 400,000 other parents.
+Why do I keep getting the same names over and over?
That is frequency weighting doing its job. About 1% of Americans are named Smith, and roughly 1 in 25 Americans carries one of just ten surnames, so realistic generation repeats common names on purpose. If you want variety, filter by a rare starting letter like Q, X, or Z, or regenerate the batch and lock in the names you already like.
+How many names can I generate at once?
As many as your browser can render. Batches of 100 to 500 names come back near-instantly because the entire name pool lives in the page itself. For a full roster of fifty NPCs or a class list of thirty, generate in rounds and copy each batch out as you go.
+Do the generators handle names with accents and diacritics?
The word banks include names like José, Zoë, and Müller where they occur naturally, but most output is plain ASCII on purpose — usernames, databases, and game engines frequently break on accented characters. If you need the accented form for fiction or genealogy, generate the base name and re-add the diacritic, or browse the French and Japanese surname pages which preserve original spellings.
+What is onomastics, and do I need to care?
Onomastics is the study of names — where they come from, how they spread, and what they signal. You do not need it to click a generator button, but it explains why top-ten baby names cover fewer than 1% of births each year, why surnames cluster by trade and place, and why the same name can spike in popularity twenty years after a soap opera character made it famous.
+Are any names off-limits or filtered out?
Real census surnames are public data and include everything the census records. Curated word banks are hand-checked, and obviously offensive strings are excluded. Generated names can still coincide with real people — that is unavoidable with any name — so avoid attaching generated names to real addresses, phone numbers, or other identifying details.