Random Binary Numbers Generator

Random binary numbers are surprisingly often what you actually want, even when the rest of your project is in decimal or hex. Generated test data, simulated sensor readings, deliberately unpredictable tokens, and a dozen debugging tricks all rely on a quick way to spin up a list of strings made of 0s and 1s. The catch is that "random" is not a single thing — uniform or biased, fixed or variable width, seeded or fresh — and the wrong choice quietly gives you boring test cases or values that do not behave like the real data you are simulating.

Random Binary Numbers Generator produces strings of any length from a single bit to over a thousand, with optional uniformity bias, optional reproducibility seed, and preset widths for the common byte, half-word and word sizes. Set the length and the count, copy the result or download it, and you are ready to paste it into a test harness, a simulation or a stress-test for a parser.

Underneath the buttons is the small amount of information the system actually requires — what uniform and biased mean, what bit-width affects, what a seed does — plus practical guidance on when to pick which. Everything runs in your browser; nothing is uploaded.

• Any length, any count: One bit to 1,024 bits, one string to 10,000 strings, with the cap adjustable up to that limit in either dimension.
• Preset widths: 8-bit, 16-bit, 32-bit, 64-bit, plus any custom width from 1 to 1,024.
• Uniform or biased: Set the probability of a 1 being emitted per bit, anywhere from 0 to 1. Uniform is the default.
• Optional seed: Reproducible sequences for testing and debugging — same seed, same output every time.
• Display density: Group bits into nibbles, bytes, words or leave them grouped only by length.
• Decimal equivalent column: Show the integer value alongside the binary string for sanity-checking the range.
• Bulk conversion utilities: Convert any list of bits you already have through the same display rules.
• Three export formats: CSV (binary only), JSON array and a plain text list with both the index and the value.
• Copy or download: Take what you generate without leaving the browser, with nothing uploaded to any server.

1. Pick a width. Use a preset (8, 16, 32, 64) or set a custom bit count from 1 to 1,024.
2. Set the count. How many values to generate.
3. Choose the bias. 0.5 is uniform; adjust up or down if you want more 0s or more 1s.
4. Set a seed if reproducibility matters. Empty for a fresh sequence on every run; a non-empty value produces the same list every time.
5. Pick a display style. Continuous bits, nibble-grouped, byte-grouped or word-grouped.
6. Toggle the decimal equivalent column if you need it. Helpful for verifying the range behaves as expected.
7. Copy or download the result. CSV for spreadsheets, JSON for code, plain text for documentation.

Example 1 — Test data for a parser. Generate 100 random 8-bit binary strings and feed them into a parser test harness. Use a fixed seed so a failing test case is reproducible across runs.

Example 2 — Simulating a noisy sensor. Generate 1,000 16-bit values with a bias of 0.05 — most values near zero with rare spikes. The result approximates a sensor that reports near-zero readings most of the time and only occasionally registers something significant.

Example 3 — Stress-testing a token generator. Set the length to 256 bits and the count to 50. Use the JSON export to produce an array suitable for fuzz-testing a token collision checker.

Example 4 — Verifying range behaviour. Use the 16-bit preset with the decimal-equivalent column enabled. The decimal column runs from 0 to 65,535, evenly distributed, which is exactly the range a 16-bit unsigned integer should reach.

Example 5 — Reversible debug aid. Pin a failing test case by switching to a deterministic seed and replaying. The exact same bits appear in the exact same order, which makes regex or bitmask logic bugs trivial to reproduce.

• Stop hand-rolling test data: Get thousands of binary strings in one click, in the width and bias you actually need.
• Reproduce any failure: A seed pins the output for debugging — re-run with the same seed and the same bits appear in the same order.
• Stay in any width: Preset widths for the common 8, 16, 32 and 64-bit cases, plus any custom width you need.
• Bias the distribution: Uniform by default, or tilt the probability of a 1 anywhere from 0 to 1 to match what real binary data looks like.
• Export the shape you need: CSV for spreadsheets, JSON for code, plain text for documentation.
• Free and private: No account, no watermark, nothing uploaded.