Your screen is too small. Please widen your browser window or use a device with a screen width of at least 300px to use this app.
Answers about Fake API: what it costs, how long your mock endpoints live, which limits apply, whether you can call them from CI, and how AI generated responses work.
Yes. Creating an account costs nothing and includes 5 projects, 10 endpoint groups per project, and 10 endpoints per group. AI generated responses are included as well, at 30 designs per day.
No. Endpoints are created in the browser and answer over HTTPS straight away. There is no CLI, no SDK, and nothing to add to your project. Your app only needs the URL.
An endpoint stays exactly as you left it until you change or delete it. Nothing expires on a timer, so a URL pasted into a ticket last month still answers today.
Yes. The URLs are ordinary public HTTPS endpoints with no authentication, so curl, fetch, Postman, Playwright, and any CI runner can call them the same way your app does.
Yes. Every endpoint has its own status code, anywhere from 100 to 599. Point one endpoint at a 500 and leave it there, and reproducing an error path no longer means editing code. At 204 the body is sent empty, as the spec requires.
Yes. Each endpoint carries its own delay, from 0 up to 60,000 ms, which is how you get your loading states and timeout branches actually exercised.
Up to 30,000 characters, with at most 50 items in any single list. That is enough for a full page of results rather than one sample row.
You write the JSON body once and tick the fields that should change. The shape of the response stays fixed, and only the ticked values are regenerated on each call, so your UI meets varied data instead of the same placeholder row every time. The feature is off by default.
Your projects and their definitions are scoped to your account and are not listed publicly. The generated endpoint URLs themselves are public and unauthenticated by design, so treat them as shareable and do not put anything sensitive in a response body.
No, and it is not meant to be. There is no persistence between calls, so a POST does not change what a later GET returns. It exists to unblock frontend work while the real API is being built.
Looking for step by step instructions instead? The documentation walks through creating a project, defining endpoints, and calling them from your app. For the wider picture, the mock API generator page covers how endpoints are defined, and the fake JSON API page covers what they can return.