Nested JSON Generator

Runs in browser Test Data

Control nesting depth, keys per level, array sizes and value types to produce complex JSON that exercises recursion limits, tree viewers and serialisers.

TEST DATA ONLY — all generated names, contacts, identifiers and numbers are synthetic and do not belong to real people or accounts.
Privacy: This tool runs entirely in your browser. Your input never leaves your device.

How to use Nested JSON Generator

  1. Set depth, keys per object and items per array.
  2. Choose the leaf types and key style.
  3. Press Generate; explore the tree or download the JSON.

Nested JSON Generator features

  • Depth up to 64 levels, up to 50 keys per object and 100 items per array
  • Arrays mixed into the structure or objects only
  • Leaf types: string, number, boolean, null, date, uuid
  • Realistic, short or very long key names
  • Tree view and JSON output with node count, depth and size
  • 200,000-node safety limit with clear messages

Nested JSON Generator example

Small nested document

Input:

Depth: 3 · Keys: 2 · Arrays: off · Leaf types: number

Output:

{
  "k0": { "k0": { "k0": 42, "k1": -7 }, "k1": { "k0": 3.14, "k1": 0 } },
  "k1": { "k0": { "k0": 9, "k1": 12 }, "k1": { "k0": -1, "k1": 5 } }
}

Frequently asked questions about Nested JSON Generator

What can I control?

Depth, keys per object, items per array, whether arrays are mixed in, leaf value types (string, number, boolean, null, date, uuid) and the key naming style.

Is there a size limit?

Yes — 200,000 nodes. Combinations that would explode (for example depth 30 × breadth 20) are rejected up front with a clear message.

What is it useful for?

Stress-testing parsers, recursion limits, schema validators, tree viewers, diff tools and serialisers with deeply nested documents.

Do all branches reach the requested depth, and is the root always an object?

Yes. The root is always an object and every branch descends to exactly the chosen depth, so the document is a full tree and the reported max depth equals your setting unless the 200,000-node cap cuts it short. Below the root, about 30% of containers become arrays when mixing is on (an array with 0 items ends its branch early).