Realistic synthetic JSON data, on demand.
JsonFabrica is an API-first platform for generating realistic, synthetic JSON test data on demand. Describe the shape of the data you need once, as a reusable template, and generate as many fresh, realistic JSON documents as you need — via a simple API call, without ever touching real customer information.
Getting realistic test data right is harder than it sounds
Fixtures go stale
Hand-written fixtures go stale, don’t scale, and rarely cover edge cases.
Real data is a privacy risk
Using real customer data for testing or demos raises privacy and compliance concerns.
Ad-hoc scripts are inconsistent
Ad-hoc "random data" scripts are inconsistent across a team, hard to reproduce, and don’t understand relationships between records.
How it works
Describe a template
Write a JSON document with placeholders that call a data-generation function, using angle-bracket syntax: <functionName(args)>.
Call the API
Save the template and call generate, or send a template body directly to the ad-hoc generate endpoint — authenticated with your API key.
Get reproducible JSON
Every response includes the seed used for that run. Reuse it to get the exact same output again, or omit it for fresh random data.
Placeholders call a data-generation function using angle brackets: <functionName(arg1, arg2, ...)>
{ "name": "<getRandomFullName()>", "id": "<createSeq('userId')>" }Key capabilities
- Template-driven JSON generation via a straightforward REST API.
- A library of roughly 25 built-in data-generation functions — names, addresses, numbers, booleans, dates, text formatting, and sequence/counter generation.
- Batch and relational generation, so related documents in the same batch can share consistent identifiers and context.
- Reproducible output through explicit or auto-generated seeds.
- Self-serve signup with instant API-key issuance — no sales call required to get started.
- Usage-based billing, so you pay for what you actually generate.
Who it’s for
Developers
Realistic fixtures and seed data for local development and automated tests.
QA & test engineering
Large, varied, repeatable datasets for functional, load, and regression testing.
Demo & sales engineers
Believable-looking sample data for product demos and sandboxes without exposing real customer records.
Data teams
Seeded, reproducible datasets to validate pipelines, analytics, or ML workflows before real data is available.
One call, one document
A single template mixing several built-in functions — a random name, email, age, and boolean flag — generated with one REST call.
curl -s -X POST http://localhost:4000/v1/templates/generate \
-H 'Authorization: Bearer sk_...' \
-H 'content-type: application/json' \
-d '{
"body": "{ \"name\": \"<getRandomFullName()>\", \"email\": \"<getRandomEmail()>\", \"age\": <getRandomNumber(18, 65)>, \"active\": <getRandomBoolean(0.8)> }"
}'{
"data": {
"name": "Jordan Blake",
"email": "jordan.blake@example.com",
"age": 34,
"active": true
},
"meta": {
"seed": 12345,
"generatedAt": "2026-08-08T00:00:00.000Z"
}
}Usage-based pricing, from startup to enterprise
Pay for what you actually generate — no per-seat pricing. Start with a 30-day free trial of Starter.