A 2Captcha alternative

Most people who leave a captcha service do it for one of three reasons: the price per solve, the free trial being too small to test with, or needing a type the service does not cover well. This page sets out what we do, how it differs, and what it takes to move an existing integration across.

What is the same

The request shape: a key, a captcha type in the path, and an asynchronous job that you poll for a result
The authentication style, using the key in a Basic authorization header so existing header code keeps working
The error model with numeric codes, including a distinct code for a job that is not finished yet
Token endpoints for reCAPTCHA, hCaptcha, Turnstile, FunCaptcha and text captchas

What is different

Sign-up includes 200 free solves, so a full integration can be validated before paying
Pricing is per solve with the cheapest types at one credit, including text captchas and Turnstile
The browser extension and the API draw on the same allowance, so a manual pass and a script are not billed through separate products
Refused submissions are not charged, and polling never costs anything

Supported types

reCAPTCHA v2 and v3, hCaptcha, Cloudflare Turnstile, Arkose Labs FunCaptcha including the matching variant, AWS WAF captcha, GeeTest, Lemin, PerimeterX and text or arithmetic captchas. Types we do not cover are not advertised, which is deliberate: a longer list that half works is worse than a short one that does.

How a migration normally goes

Create an account and copy the key from the dashboard
Point the base URL at our host and keep your request bodies unchanged
Switch the key prefix from whatever you use today to the one we issue
Run one solve per captcha type you rely on and check the response shape against your parser
Remove the old client once the types you need are confirmed

What to check before switching

Whether the types you actually hit are on the supported list, including edge cases such as matching FunCaptcha
Whether the errors you see most often have a distinct code, so retry logic can tell pending from terminal
Whether failed solves are billed, because that dominates cost in a noisy scraping pipeline
Whether you can test at realistic volume without paying, which is what the free tier is for

The honest comparison

Very large operations with decades of accumulated traffic may still be better served by the incumbent. Where we fit is teams that want the same asynchronous API shape, a trial large enough to validate against, per-solve pricing with the cheap types priced cheaply, and a supported type list that is short and real rather than long and partial.

How to integrate

curl -X POST https://fuckcaptcha.top/v1/token/recaptcha2 \\
  -H "Authorization: Basic $NSK_KEY" \\
  -d '{"sitekey":"<sitekey>","url":"<page url>"}'

Pricing

Charged per solve: one credit for each recognition request, and polling for the result is free. Signing up includes 200 free solves.

FAQ

Can I keep my existing code?

Mostly. The flow of submitting a job and polling for a result is the same, so usually the base URL and the key change and the request bodies stay as they are.

How many free solves do I get?

200 on sign-up. That is enough to run every type you depend on, including token types, and verify the response handling before spending anything.

Which types are cheapest?

Image recognition and Turnstile cost one credit. Token solves for reCAPTCHA and hCaptcha cost substantially more, which mirrors how the upstream is priced.

Is there a minimum commitment or subscription?

No. Balances are bought and consumed per solve, so a pipeline that runs for a week and then stops does not keep billing.

What if a type I need is not listed?

Ask directly before migrating. The supported list on this site is the accurate one, and it is better to find out before rewriting an integration.

Read the API reference Download the Chrome extension