Home Fake data API

A free fake data API whose answers never change

Fetch synthetic users, products and orders as static JSON with no key and no sign-up. Every file comes from a fixed seed and stays the same once published, so your tests can assert against it.

Endpoints

Every endpoint is a plain GET on a static file. No key, no sign-up, no rate-limit header to watch, and CORS is open, so a browser test can fetch it from any origin.

PathReturns
/api/v1/users.json100 users. Fictional people with coherent name, email and gender, and a matching city and country.
/api/v1/users/{id}.jsonOne record, id 1–100.
/api/v1/products.json100 products. A small catalogue: name, SKU, category, price and stock flag.
/api/v1/products/{id}.jsonOne record, id 1–100.
/api/v1/orders.json100 orders. Orders whose userId and productId point at records in the two collections above.
/api/v1/orders/{id}.jsonOne record, id 1–100.
/api/v1/reserved-test-values.jsonReserved domains, documentation IP ranges and provider test card numbers, each with its source.
/api/types.jsonThe catalogue of all field types the generator offers.
/openapi.jsonAn OpenAPI 3.1 description of everything above.

Fetch it from a test

From a shell:

curl https://fundata.dev/api/v1/users/1.json

From JavaScript, in a browser or Node 18+:

const users = await fetch('https://fundata.dev/api/v1/users.json').then(r => r.json());
console.log(users.length); // 100

From Python:

import json, urllib.request
with urllib.request.urlopen('https://fundata.dev/api/v1/orders/7.json') as r:
    order = json.load(r)
print(order['userId'], order['status'])

The first user, exactly as the API returns it:

{
  "id": 1,
  "firstName": "Anthony",
  "lastName": "Khan",
  "email": "anthonykhan@example.org",
  "gender": "Prefer not to say",
  "username": "brave_badger576",
  "phone": "+1 (342) 555-0144",
  "company": "Pinnacle Solutions",
  "jobTitle": "Customer Success Manager",
  "city": "Kuala Lumpur",
  "country": "Malaysia",
  "registeredOn": "2025-09-14"
}

What is in each collection

users — Fictional people with coherent name, email and gender, and a matching city and country. Fields: id, firstName, lastName, email, gender, username, phone, company, jobTitle, city, country, registeredOn.

products — A small catalogue: name, SKU, category, price and stock flag. Fields: id, name, sku, category, price, inStock.

orders — Orders whose userId and productId point at records in the two collections above. Fields: id, userId, productId, quantity, status, orderedOn.

In orders, userId and productId are always between 1 and 100, so every order joins to a user and a product. Names, emails and genders agree within a record, and each city belongs to its country. Dates have fixed bounds and no values depend on the day they were generated.

The bytes do not change

An assertion written against /api/v1/users/1.json today will still pass next year. The files are generated once from fixed seeds through the same engine as the test data generator, and the build refuses to overwrite a published /v1/ file. If the data ever has to change, it will appear under a new /v2/ path, and /v1/ will stay as it is.

That makes these files useful for snapshot tests and documentation examples. It also means they are small and fixed: 100 records per collection, no filtering, no paging.

Reserved values with their sources

reserved-test-values.json lists values that are set aside so they cannot belong to anyone. It includes the example domains from RFC 2606, the .localhost name from RFC 6761, the IPv4 documentation ranges from RFC 5737, and the IPv6 documentation prefix from RFC 3849. It also lists card numbers that Stripe publishes for its test mode. Each entry carries the link to its source document, so a reviewer can check it.

When to use something else

This API is a set of fixed files, so it has limits. For data shaped to your own schema, a different row count or a different seed, use the generator in the browser, or the local CLI. For an endpoint that accepts POST requests and echoes them back, a local mock server fits better; the API mock data guide covers that setup. And for anything a production system will read, generate fresh data and review it before you rely on it.

Machine-readable description

/openapi.json describes every endpoint in OpenAPI 3.1. The record shapes are JSON Schema, produced by the same code as the generator's JSON Schema export. Import it into an API client to get the requests ready-made, or give it to a code generator to get typed models.

Last updated