This free random number generator gives you unpredictable numbers within any range you choose, one at a time or by the hundred, with control over how many you get, whether repeats are allowed, and how the results are sorted. Set a minimum and a maximum, pick how many numbers you want, decide whether duplicates are permitted, and generate. The result appears at once, worked out on your own device, so nothing you enter and nothing you get back is uploaded or stored. Use it to run a raffle, pick lottery numbers, draw a fair sample for research, decide who goes first, or simply get an honest number when the one in your head would not really be random at all.
Getting a genuinely random number is harder than it sounds, and it is something people are famously bad at doing by themselves. Ask a room of people to name a number from one to ten and far too many will say seven, while almost nobody picks one or ten. Our minds carry biases we do not notice, which is exactly why a proper random number generator is useful: it has no favorites, no habits, and no way to be nudged. This page explains what the tool does, how to use its options well, where random numbers matter in real life, and the honest truth about what kind of randomness a browser can and cannot provide.
What a random number generator is, and when you need one
A random number generator, often shortened to RNG, is a tool that produces numbers you cannot predict, drawn from a range you define. At its simplest it hands you one number between two bounds. With a few options it can hand you a whole list, guarantee that none repeat, and sort them for you.
The situations that call for one are more varied than you might guess. A raffle or giveaway needs a fair way to pick winners from numbered entries. A lottery player wants a random line rather than the same birthday numbers everyone else uses. A researcher needs to select a sample and assign people to groups without bias, which is the backbone of a credible study. A teacher wants to call on students at random or generate data for a lesson. A game needs a random result. A team needs to decide an order of turns. And plenty of everyday moments, from choosing which of thirty restaurants to try to picking a seat, come down to wanting one honest random number.
In each case the value of the tool is the same: it removes human bias and gives every possible outcome an equal, unpredictable chance.
How the ToolFiddle random number generator works
When you press generate, the tool asks your browser's secure random source, a function called crypto.getRandomValues, for raw random values and maps them onto the range you set. If you asked for a number between 1 and 100, it spreads the raw randomness evenly across those hundred possibilities so that each one, from 1 to 100, has the same chance of coming up. If you asked for a list, it repeats the process for each number, applying your choices about duplicates and sorting as it goes.
One subtle detail is worth spelling out, because it separates a careful generator from a sloppy one. The raw values a browser hands back are huge numbers, and the obvious shortcut for squeezing them into your range, taking the remainder after division, quietly favors the lower part of the range whenever the range does not divide the raw span evenly. This flaw is called modulo bias. Our generator avoids it with a technique called rejection sampling: any raw value that falls in the uneven leftover region is simply thrown away and redrawn, so the values that survive map onto your range perfectly evenly. The cost is an occasional extra draw you will never notice, and the benefit is that every number from your minimum to your maximum has exactly the same chance, not just approximately the same. That is what makes the draw unbiased in the strict sense, not only in spirit.
Every draw is independent unless you tell the tool otherwise by forbidding repeats. That means, with duplicates allowed, getting a 42 on one draw does nothing to change the chance of a 42 on the next. There is no memory and no attempt to balance the results, because true randomness does not work that way. The evenness you expect from a fair generator shows up over many draws as each number appears about equally often, not by the tool steering individual results.
The whole calculation happens on your device, in your browser, which has two consequences worth stating plainly. The results are instant, because nothing waits on a server, and they are private, because nothing is sent anywhere. More on both below.
How to use the random number generator step by step
The tool is designed so you can get a number within a second of the page loading, then refine the options only if you need to.
- Set your range. Enter the minimum, the smallest number you want, and the maximum, the largest. For a number between 1 and 100, that is 1 and 100. The minimum can be negative if your range needs it.
- Choose how many numbers. For a single value, leave it at one. For a list, enter the count you want, whether that is six for a lottery line or a few hundred for a data set.
- Decide on duplicates. Leave repeats allowed when each draw should be independent. Turn on no repeats, or no duplicates, when every number in the list must be unique, such as in a raffle draw or a sample.
- Pick a sort order. Sort the results into ascending order to read them easily, or keep them in draw order when the sequence itself is the answer, such as an order of play.
- Filter odd or even if you need to. The allowed numbers switch limits the draw to odd values only or even values only, which saves fiddling with the range when only one parity qualifies.
- Generate. Press the button and your number or list appears immediately, with earlier draws kept in a history underneath so nothing is lost. Generate again as often as you like, with no limit, because the work is done on your device.
The options that make it powerful
The difference between a toy and a genuinely useful random number generator is in the options, so it helps to understand what each one is for.
The range, set by the minimum and maximum, defines the pool of possible results. Every whole number from the minimum to the maximum, inclusive, has an equal chance. Widening the range spreads the possibilities; narrowing it concentrates them.
The count controls how many numbers you get in one generation. This turns the tool from a single picker into a batch generator, which is what you need for sampling, multi winner draws, or generating test data.
The duplicates setting is the one people overlook, and it changes the meaning of your draw entirely. Allowing repeats treats each pick as independent, like rolling the same die over and over, so the same number can appear more than once. Forbidding repeats draws without replacement, like pulling numbered tickets from a hat and not putting them back, so every result is unique. Choosing the right one is essential to a fair result, and the next sections show why.
The sort option decides how the list is presented. Sorting into order suits lottery lines and tidy samples. Keeping draw order suits any case where the order carries meaning, such as deciding first, second, and third place.
The odd or even filter restricts the pool to numbers of one parity. Set it to odd and a range of 1 to 100 draws only from the fifty odd values; set it to even and only the even ones qualify. It is a small option that comes in handy more often than you would think, from splitting entries into two groups to matching numbered seats or house rules in a game.
A worked example: drawing raffle winners
Suppose you have run a charity raffle and sold 250 tickets, numbered 1 to 250, and you want to draw three winners fairly. This is a textbook job for the random number generator with no repeats.
Set the minimum to 1 and the maximum to 250, so the pool matches your tickets. Set the count to three, because you want three winners. Turn on no repeats, because the same ticket must not win twice. Then generate. The tool returns three unique numbers, say 47, 183, and 6, and those are your winners.
What makes this defensible is that the draw is random, on your device, and impossible to influence. Nobody's ticket had a better or worse chance, every number from 1 to 250 was equally likely, and no one, including you, could steer the outcome. If you wanted to draw the winners in a set order, first prize then second then third, you would keep the results in draw order rather than sorting them, so the first number drawn takes the top prize. And if your entries are names rather than numbers, the random picker draws winners straight from a pasted list without the numbering step.
A worked example: random sampling and assigning order
Research and data work lean on random numbers constantly, and a second example shows a different use of the same options. Say you have a list of 500 survey responses and you want to check a random sample of 30 of them for quality. Number the responses 1 to 500, set the range to 1 and 500, set the count to 30, and turn on no repeats so you do not check the same response twice. The tool returns 30 unique numbers, and those are the responses you inspect. Because the selection is random, your sample is not skewed toward the start of the list, the most recent entries, or any group, which is what makes findings from a sample trustworthy.
Assigning order works the same way. To decide the running order for eight presenters, number them 1 to 8, generate all eight with no repeats, and keep the draw order. The sequence you get is a fair, random order of play that nobody arranged. This kind of random assignment, where chance rather than choice decides who goes where, is exactly why random number generators sit at the heart of fair experiments and fair competitions alike.
Real world uses across different people
A random number generator earns its keep across a wide mix of people and tasks.
For anyone running a draw, it is the fair broker. Raffles, giveaways, prize draws, and door prizes all come down to picking numbered entries at random, and doing it with a generator that cannot be nudged makes the result easy to trust and easy to defend.
For lottery and game players, it supplies a random line free of the human habit of clustering numbers around birthdays and lucky figures. It cannot change the odds, since every combination is equally likely, but it gives a genuinely spread pick and settles any in game moment that needs a random value.
For researchers, statisticians, and students, it is a core instrument. Random sampling and random assignment are what separate a credible study from a biased one, and a generator with a no repeats option and a wide range does both cleanly. It also produces random data sets for practicing analysis.
For teachers, it is a fair way to call on students, form random groups, or assign an order of presentations, taking any accusation of favoritism off the table. A spinning wheel of names does the same job with more theater when the class is watching. For event organizers and workplaces, it settles orders of turns, seat assignments, and who gets first pick. And for everyday decisions, it turns a stubborn choice among numbered options into a quick, honest result.
The kind of randomness you are getting, honestly
It is worth being straight about what kind of randomness this tool provides, because plenty of sites gloss over it. The numbers come from crypto.getRandomValues, the cryptographically secure random generator built into your browser. That is the same source the browser trusts for its own security work, seeded with unpredictable entropy your operating system gathers from the physical world, and it is a clear step up from the basic Math.random function that many casual tools lean on. Strictly speaking, even this is not a true hardware generator: the values are produced by an algorithm rather than read directly off a physical noise source like atmospheric static or radioactive decay, which is what laboratory grade equipment uses.
For the uses this tool is built for, that is more than enough. The output is thoroughly unpredictable to any person, evenly spread across your range thanks to the rejection sampling described earlier, and free of patterns you could exploit in a raffle, a game, a sample, or a draw. When you generate a number between 1 and 100, you have no way of knowing what it will be, and every value is exactly equally likely. That is all fairness requires here, with room to spare.
Where the distinction between grades of randomness really matters is secrets. Passwords, encryption keys, and security tokens must resist an expert who studies past outputs and tries to predict the next one, and they need careful construction well beyond drawing a number. Although this generator draws from the secure source, it is a number picker, not a security tool. If you need a strong password, use the password generator, which is built for exactly that job end to end. We would rather tell you that plainly than pretend a number picker is something it is not.
Randomness with and without replacement
One idea underlies the duplicates option and deserves a section of its own, because getting it wrong quietly breaks a draw. Statisticians talk about sampling with replacement and without replacement, and it is simpler than it sounds.
Sampling with replacement is like drawing a numbered ball from a bag, noting it, and putting it back before the next draw. The same number can come up again, and every draw faces the same full set of possibilities. In the tool, this is repeats allowed. It suits anything where each pick is meant to be independent, such as simulating many rolls of one die, which the dice roller also does with proper faces, generating test data where duplicates are fine, or any process where the same value recurring is natural.
Sampling without replacement is like drawing balls and setting them aside, so no number can appear twice and the pool shrinks with each draw. In the tool, this is no repeats. It suits anything where each result must be distinct: raffle winners, lottery numbers, a sample of unique items, or an order of play. Getting this setting right is the difference between a fair draw and a broken one. If you drew raffle winners with repeats allowed, the same ticket could win twice while another valid ticket never had its fair turn, so no repeats is the correct choice there.
A quick way to remember it: if drawing the same number twice would be a problem, turn on no repeats. If it would be harmless or even expected, leave repeats allowed.
Tips and common mistakes
- Check your range bounds. The minimum and maximum are both included, so a range of 1 to 100 can return 1 or 100. If you need 1 to 99, set the maximum to 99. Off by one range errors are the most common slip.
- Match the duplicates setting to the job. This is the mistake that matters most. Use no repeats for draws and samples where results must be unique, and allow repeats where each pick is independent. The wrong choice can make a draw unfair without it being obvious.
- Do not expect small batches to look even. Ask for ten numbers from 1 to 100 and they may cluster or leave gaps. Evenness shows up over hundreds or thousands of draws, not tens, which is the law of large numbers at work. A lumpy small sample is normal, not a fault.
- Do not treat past results as a guide. With repeats allowed, a number that has not appeared is not due, and one that appeared twice is not hot. Each independent draw ignores history entirely.
- Do not rely on it for secrets. For a raffle it is perfect. For a password or security key, use a tool built for security end to end instead, because a number picker is not designed for that threat.
Running a draw people will trust
If you are using the generator for something with a prize or a stake, a little process makes the result easy to defend. Decide and announce the rules before you draw: the exact range, how many winners, and that repeats are off so nobody can win twice. Number the entries clearly so every one maps to a value in the range. Draw once and stick with the result, rather than regenerating until you like it, since re drawing quietly undoes the fairness. If witnesses are watching, generating on screen where they can see the range and the count makes the whole thing transparent, and the history of previous draws under the result shows exactly what was drawn and when. Because the draw runs on your own device with no way for anyone, including us, to influence it, a clean process is all you need for a result people will accept.
Your numbers stay on your device
Everything the random number generator does happens inside your browser, on the device in front of you. When you press generate, your own device produces the numbers and displays them, with no request going to a server and nothing leaving your hands. We do not upload your ranges, your results, or anything else, we do not log them, and we keep no history after you close the tab. The numbers are ephemeral, and we treat them that way by storing nothing at all.
Because the work is local, the tool also runs offline. Once the page has loaded you can switch off your connection and keep generating, since none of the process depends on the internet. That is handy for a draw at a venue with poor signal, or simply for the reassurance that a tool you use for fair draws is not quietly sending your data anywhere. A random number generator should not need to know anything about you, and this one does not.
Genuinely free, with no catch
The random number generator is free in every sense. There is no sign up, no account to create, and no email to hand over. There is no watermark, no daily limit on how many numbers you can generate, and no option locked behind a paid plan. Some sites cap free use, or reserve features like bulk generation or the no repeats option for subscribers. We do not, because the generator costs almost nothing to run once it has loaded on your device, so there is no reason to charge for it or hold anything back.
That means you can draw one number today to settle a small question, run a raffle for five hundred tickets tomorrow, and generate a data set of thousands next week, all without paying, registering, or being pushed toward an upgrade. The range control, the count, the no repeats option, and sorting are available to everyone, every time.
Instant, unlimited, and light
Because the random number generator runs on your device rather than a distant server, your numbers appear the instant you press generate. There is no spinner waiting on a network round trip, no slowdown when many people use the site at once, and no artificial cap on how often you can draw. You press, and the results are there.
The page stays light and free of the heavy advertising and tracking scripts that weigh down so many free tools, so it loads quickly and responds immediately to every generation. Drawing a single number or a list of hundreds, over and over, stays fast and smooth without draining your battery or your patience.
Works on any device, even offline
The random number generator behaves the same on a laptop, a tablet, or a phone, and the layout adjusts to whatever screen you use. Any current browser runs it. Because the generation is done locally, you do not need a connection once the page has loaded, so a dropped signal at an event or on the move does not stop a draw. Nothing depends on signing in or staying online, which is how a quick, dependable utility should work.
What separates a good random number generator from a poor one
Online random number generators vary more than they look, and a few things mark out a good one. It spreads results evenly across the whole range so every value is equally likely, with no bias toward the middle or the ends. It offers a real no repeats option for draws that need unique results, and it lets you control the count and the sort order. It is honest about the kind of randomness it uses rather than overselling it. And it does not bury the generate button under ads or make you wait on a slow server.
Our approach is to do all of that plainly: an even, rejection sampled draw across your exact range from your browser's secure randomness, computed on your device, with control over how many numbers, whether repeats are allowed, and how they are sorted, plus a straight answer about what the randomness is and is not. Fair results, useful options, honesty, and speed together are what a random number generator should offer, and that is what we built.
Related tools on ToolFiddle
If a numbered range is more than you need, a couple of neighboring tools cover simpler kinds of randomness. For a plain two way choice, the coin flip gives heads or tails with a running tally. For game style dice with faces and modifiers, the dice roller rolls the full polyhedral set and totals the results. And when you need a random secret that must resist attackers rather than a number for a draw, the password generator uses cryptographically secure randomness to build strong passwords from end to end. Each of these, like the random number generator, runs entirely in your browser and keeps nothing.
The short version
A random number generator hands you unpredictable numbers within any range you set, from a single pick to a long list, and this one lets you control how many you get, whether they repeat, and how they are sorted. Set the minimum and maximum, choose your count, decide on duplicates, and generate, all on your own device. Turn on no repeats for draws and samples where results must be unique, and leave repeats allowed where each pick is independent. The randomness comes from your browser's secure generator with rejection sampling on top, so every value in your range is exactly equally likely, and we say plainly what that does and does not cover. The tool is free with no limits, stores nothing, and works offline once loaded. When you need a fair number, it is ready the instant you press generate.