10 Best Public & Dummy APIs for Testing (2026)

Table of Contents

1. Comparison Table
2. Fake CRUD Data APIs
3. HTTP Behavior Simulation
4. Real Public APIs
5. Built for Test-Automation Practice
6. How to Choose
7. FAQ

Comparison Table

Fake CRUD Data APIs

• JSONPlaceholder — the default choice used throughout most current API testing tutorials, including every framework tutorial on this site. No auth, predictable resources (/users, /posts, /comments, /albums, /photos, /todos), and stable, well-documented behavior.
• ReqRes — user-management-shaped resources with realistic pagination, useful when your test cases specifically need to exercise paginated list endpoints.
• Fake Store API — e-commerce-themed data (products, carts, users), useful for practicing test cases shaped like a shopping-cart or catalog domain rather than generic posts/comments.
• DummyJSON — a larger dataset across products, carts, users, posts, comments, quotes, and recipes, with a simulated login flow for practicing authenticated request patterns.

Important for all four: writes don't persist. A POST returns a realistic 201 with what looks like a newly created resource, but nothing is actually stored — a follow-up GET for that same ID won't return it. This is fine for learning framework syntax; it's not a substitute for testing against a real backend with real state.

HTTP Behavior Simulation

• httpbin.org — not a fake business domain, but a tool for simulating exact HTTP behaviors: /status/404 returns a specific status code on demand, /delay/3 responds after a defined number of seconds, /headers echoes back what your client actually sent, /redirect/3 chains through a defined number of redirects. This is the tool to reach for when you need to verify your own test framework's timeout, retry, or redirect-following behavior actually works — see the common pitfalls sections in our framework tutorials for why an unhandled timeout is a recurring real bug.

Real Public APIs

• PokeAPI — a real, free, no-auth public API with genuinely deep, nested response data, useful specifically for practicing schema validation and nested-field assertions against realistic complexity a flat fake-data API doesn't provide.
• GitHub REST API — real production infrastructure with generous unauthenticated rate limits (higher with a token), useful for practicing against real pagination headers, real rate-limit responses, and a genuinely well-documented, versioned API contract.
• OpenWeatherMap — a real API requiring a free-tier API key, useful specifically for practicing API-key-based authentication and handling a third-party service with its own rate limits and quota behavior.

Built for Test-Automation Practice

• restful-booker — a demo API created specifically for the test-automation community to practice against, including a token-based auth flow, making it a common choice specifically for learning authenticated request patterns beyond what the simpler fake-CRUD APIs above cover.

How to Choose

• Learning a new framework's basic syntax → JSONPlaceholder — the most predictable, most-documented default.
• Need pagination in your example test cases → ReqRes.
• Need auth-flow practice specifically → DummyJSON or restful-booker.
• Need to verify your own framework handles timeouts/retries/redirects correctly → httpbin.org.
• Need deeply nested response data for schema validation practice → PokeAPI.
• Need real rate-limiting and pagination behavior, not simulated → the GitHub REST API.

Frequently Asked Questions

What is the best free dummy API for testing? JSONPlaceholder — free, no auth, predictable fake CRUD data, used throughout most current API testing tutorials.

What is httpbin.org used for? Testing your own tooling against specific HTTP conditions — status codes, delays, redirects, header echoing — rather than simulating a fake business domain.

Do writes to a dummy API like JSONPlaceholder actually persist? No — they simulate a realistic response without actually storing the change.

What is restful-booker and why was it built? A demo API purpose-built for the test-automation community to practice against, including token-based auth.

Should I test against a real public API or a dummy one? Dummy APIs for learning a framework or writing examples; real public APIs when you want to practice against genuine rate limiting, pagination, and auth.

Can I use these APIs to test my own test framework's timeout and retry handling? Yes — httpbin.org's /delay/{n} endpoint is built exactly for this.

Key Takeaways

• JSONPlaceholder is the default for a reason — stable, documented, no-auth, and used throughout this site's own tutorials.
• Writes to fake-data APIs never persist — don't mistake a realistic 201 response for actual stored state.
• httpbin.org tests your tooling, not a fake domain — reach for it specifically to verify timeout/retry/redirect handling.
• Real public APIs (PokeAPI, GitHub) are worth using when you need genuine rate-limiting and pagination behavior, not a simulation of it.
• restful-booker exists specifically for test-automation practice, including auth flows the simpler fake-CRUD APIs don't cover.

Related Articles

• API Testing with Python: pytest + requests Tutorial — one of many tutorials on this site using JSONPlaceholder as the example API.
• How to Write API Test Cases — a template you can practice against any API on this list.
• API Mocking & Service Virtualization Tools — building your own fake API when none of these fit your specific schema.
• Types of API Testing — what to actually test once you have an API (real or dummy) to point your suite at.