SeedBase
A mock REST API whose data actually makes sense.
Build your frontend against an API the backend hasn't built yet. Unlike a flat fake-data server, SeedBase serves a referentially consistent database: /users/1/orders returns orders that really belong to user 1, every foreign key resolves, and distributions look like real life.
No backend · no login to call it · CORS open · free tier, no card
From schema to a live REST API in two minutes
Pick a template or import your own schema (SQL, Django, Prisma), generate, and enable the mock API. You get an unguessable token URL you can hit straight from the browser.
curl "https://seedba.se/api/v1/mock/<token>/users"
# → [ { "id": 1, "name": "Alex Smith", "email": "..." }, ... ]
curl "https://seedba.se/api/v1/mock/<token>/users/1/orders"
# → orders that actually belong to user 1 (FK resolves)What the mock API gives you
Relations that resolve
Follow foreign keys: /orders/5/line_items, /users/1/orders. Children point at parents that exist, so your relational UI has something real to render.
Pagination, filter, sort, search
?page=&page_size=, ?status=open, ?_sort=created_at&_order=desc, ?q=text. Plain JSON arrays with an X-Total-Count header, the convention you already know from JSON Server.
No backend, no login
Token in the URL, CORS open. Call it from your React/Vue/Svelte app on localhost without auth headers or a proxy. Disable it with one click.
Real distributions
Not every user has exactly five orders. Long-tail and normal distributions surface the pagination, empty-state and N+1 edges that flat mock data hides.
Domain-aware values
Pick an industry and product names, categories and brands match it, a medical project serves drug names, a toy shop serves toys, not generic placeholders.
Straight from your editor
Enable the mock API and copy its URL from the VS Code or JetBrains plugin, or insert FK-consistent sample rows right at your caret as SQL, PHP, Python or JSON.
FAQ
What makes this different from a normal mock API?
Tools like JSON Server, Mockoon or Mockaroo return flat, unrelated rows. SeedBase reads your schema and serves a referentially consistent database: /users/1/orders returns orders that actually belong to user 1, products point at categories that exist, distributions are realistic. Data that is structurally real, not just real-looking.
Do I need a backend or any setup?
No. Pick a schema, generate, enable the mock API. You get a token URL you can call from the browser, no login, CORS open. Built for developing a frontend against an API that doesn't exist yet.
Which query features does it support?
Pagination, exact filters, full-text search (?q=), sorting (?_sort=&_order=), single rows and foreign-key navigation. Plain JSON arrays with X-Total-Count.
Can I make the data fit my industry?
Yes, pick an industry and the product names, categories and brands match it. More on the generator →
Mock your API in two minutes
Sign up, pick a schema, enable the mock API. No backend, no credit card, the free tier is enough to wire it into your frontend today.
- FK-consistent
- Pagination & filter
- No login to call
- EU-hosted
Related: test data from SQL · Prisma · Compare: vs Mockaroo · vs Faker