Draw secure random integers instantly in your browser. Set any minimum and maximum, pick how many numbers you need, allow repeats or force uniqueness, then copy the set or download it - all processing stays on your device.
Technical Specifications
Entropy source and drawing rules used for every set.
| Specification |
Details |
| Entropy source |
Web Crypto API (crypto.getRandomValues) with rejection sampling - every integer in range has equal probability |
| Supported range |
Whole numbers from -1,000,000,000 to 1,000,000,000; both endpoints inclusive |
| Batch size |
1 to 1,000 numbers per set, each drawn independently |
| Uniqueness |
Optional no-repeat mode; quantity must fit inside the range size or the draw is rejected with a message |
| Ordering |
Draw order, ascending, or descending - applies to display, copy, history, and download together |
| Statistics |
Live minimum, maximum, sum, and average for every set, shown in the breakdown card |
| Processing location |
100% client-side JavaScript in your browser memory; nothing is uploaded |
| Export |
Copy full set to clipboard, copy any recent set, or download a plain-text report |
| Usage limit |
None - draw unlimited sets for free, no signup required |
Understanding Random Number Draws
A fair draw means each candidate number has the same probability as every other. Computers normally produce deterministic pseudo-random sequences, which are fine for games but predictable in principle. This tool instead samples entropy from your operating system through the Web Crypto API, the same source browsers use for security keys, so draws cannot be predicted or replayed.
Why rejection sampling matters
A common shortcut maps a large random value onto a range with a remainder operation, but unless the range divides the source evenly, low numbers come up slightly more often. Rejection sampling discards the few values that would cause this skew and draws again, keeping every integer in your range exactly equally likely - whether the range holds 6 faces or 200 million tickets.
Unique draws and range sizing
With repeats allowed, each draw is independent: rolling 1 to 6 ten times can absolutely show the same face twice. Unique mode instead deals numbers like cards from a deck, so a range of 1 to 50 yields at most 50 distinct picks. If you request more unique numbers than the range holds, the tool stops you with a message rather than silently repeating. For database keys and identifiers rather than picks, the UUID Generator is the better fit.
When to use which settings
Use 1 to 6 or 1 to 20 for dice-style game rolls, 1 to 100 for percentage checks, and a narrow unique range like 1 to 30 for classroom pickers or giveaway winners. Large batches up to 1,000 suit simulations and sampling, where the live sum and average confirm the distribution at a glance. For secret codes and logins instead of draws, switch to the Password Generator.
Honest limitations
This tool draws whole numbers only - it does not produce decimals, fractions, weighted outcomes, or shuffled text lists. Ranges are capped at plus or minus one billion and batches at 1,000 numbers to keep the page instant. Results are random, not lucky: past draws never influence future ones, so no pattern of previous sets predicts the next.
Privacy and Client-Side Security
Privacy Guarantee: All generation occurs locally inside your browser. Your ranges, settings, and results are never transmitted, stored, or recorded on remote servers.
Many online pickers log draws or fetch numbers from a server, which exposes raffles, selections, and game outcomes to a third party. TinyToolzz keeps everything in-page: entropy sampling, range mapping, sorting, and statistics run with plain JavaScript in your tab's memory.
Because nothing is uploaded, there is no waiting, no account, and no copy of your numbers retained anywhere. Your values exist only in the tab's memory and are discarded when you close the page or click Clear.