Seeded random number generator in a range

Supply the same range, seed, and mode later and you receive exactly the same result. That reliable reproducibility makes it useful for test fixtures, simulations, games, demonstrations, and any workflow where an apparently random choice must be replayed or audited. It uses no system randomness, network service, or current time.

Generate a number that you can reproduce

Ordinary random-number tools are convenient until you need to explain or repeat a result. This generator treats the seed as part of the input, so a generated value becomes a reproducible fact rather than a transient accident. Enter an inclusive minimum, an inclusive maximum, and a safe-integer seed. Keep the default integer mode when you need a whole number, such as an index, score, quantity, or test case. Select float mode when a fractional value is appropriate. The response repeats the effective bounds, seed, and mode alongside the generated value, which makes saved output self-describing. If you send the same four inputs from a browser, script, CI job, or API client, the calculation follows exactly the same fixed arithmetic path. This property is especially helpful when a failure depends on a generated input: record the seed with the test report, rerun it later, and inspect the identical case without preserving hidden generator state.

Understand inclusive bounds and validation

Both bounds are inclusive. In integer mode, a range from 1 through 6 can return 1, 6, or any whole number between them. Integer mode requires the minimum and maximum to be safe integers, because JavaScript numbers outside that exact range can silently lose units and make a promised boundary misleading. Float mode accepts any finite numeric bounds and derives a reproducible fraction from 53 mixed seed bits; its normalization permits both endpoints. A range whose minimum equals its maximum simply returns that shared value in either mode. The generator rejects missing values, non-finite values such as Infinity, unsafe seeds, unsupported modes, and integer-mode bounds that contain fractions. It also reports an input error when the minimum exceeds the maximum, instead of silently swapping the values. Explicit rejection is important in automation: reversed configuration should fail visibly rather than produce plausible output from a range the caller did not intend.

Use seeded randomness for repeatability, not security

A deterministic generator is ideal when repeatability is the goal. Use it to choose a stable example for documentation, distribute fixtures across a bounded set, create replayable game scenarios, or feed controlled variation into a simulation. Different seeds generally produce different mixed values, while reusing a seed preserves the result for the same range and mode. The algorithm performs a fixed 64-bit integer mix locally and never reads the clock, operating-system entropy, external storage, or a remote service. That independence keeps results stable and makes each request fast. It also means this capability is deliberately not a security generator. Anyone who knows the algorithm and seed can reproduce the output, so do not use it for passwords, authentication tokens, encryption keys, lottery drawings, regulated selections, or adversarial games. For those jobs, choose a cryptographically secure entropy source with the appropriate operational controls. Here, the seed is a replay key, not a secret.