What score will I get?

Score is decided by Google's backend when your server calls siteverify. The biggest driver is the IP that fetches /recaptcha/api2/* — the ProxyLess path typically scores 0.3-0.5; for 0.7-0.9 use ReCaptchaV3Task with your own proxy. We can't guarantee a specific score because we never see what Google returns to your siteverify call.

Do I need to provide the pageAction?

It's recommended. The pageAction must match what the site's JavaScript passes to grecaptcha.execute(). If it doesn't match, the score will be lower.

What's the difference between v3 and v3 Enterprise?

reCAPTCHA v3 Enterprise uses a different API endpoint and may include additional risk signals. Set isEnterprise: true if the site loads from recaptcha/enterprise.js.

Can I report my verified score back to you?

Yes — call POST /reportScore with the score Google returned from siteverify. It's optional, never affects pricing, and powers the V3 quality dashboard so you can see your score distribution over time.

How do I find the sitekey?

Search the page source for 'data-sitekey', 'sitekey', or the pattern starting with '6L'. It's also visible in network requests to google.com/recaptcha.

Can I use this with Selenium/Puppeteer?

Yes. Create the task via our API, get the token, then inject it into the page using document.getElementById('g-recaptcha-response').value = token.

Is there a rate limit?

No hard rate limit. Our infrastructure auto-scales. You can send thousands of concurrent tasks.

Before you submit — these must be correct

If any of these are wrong, your token may be accepted by our solver but rejected or scored poorly by the target site.

pageAction must match the site's real action. Google scores the token against the action you request — if the site calls grecaptcha.execute(sitekey, {action: 'login'}) and you send pageAction: 'verify', the score will drop. Inspect the site's JS for the exact string. We default to 'verify' only when none is provided.
websiteURL must be the exact page URL where reCAPTCHA loads. Domain-only URLs or wrong paths cause invalid tokens — use the full URL of the page that contains the widget.
Proxy quality is the #1 factor in your score. Google evaluates the IP that fetches /recaptcha/api2/*. The ProxyLess path produces lower scores than a dedicated, high-reputation IP you supply yourself. For 0.7-0.9 scores on strict sitekeys, use ReCaptchaV3Task with your own proxy.
isEnterprise must match the site. Enterprise v3 uses a different bundle (enterprise.js) and scoring endpoint. Setting it wrong will produce an invalid token.

reCAPTCHA v3

ReCaptchaV3TaskProxyLess

reCAPTCHA v3 Solver, solved in 5.0s.

Bypass Google's invisible reCAPTCHA v3 with AI-powered token generation that passes server-side verification.

Get API Key Try Demo Examples

Avg Speed~5.0s

Success99%+

Cost / 1k$2.00

Throughput12/m

reCAPTCHA v3 is Google's invisible captcha that runs in the background without user interaction. Instead of showing a challenge, it analyzes user behavior and returns a risk score between 0.0 (bot) and 1.0 (human). Websites set a threshold (typically 0.3-0.7) and block requests below it. Unlike v2, there's no checkbox or image puzzle — the captcha is completely invisible. This makes it harder to detect but also harder to bypass, since you need to produce a token that results in a high enough score.

Quick Integration

solve.py

import requests, time

API = "https://api.capzy.ai"
KEY = "capzy_your_key_here"

# Step 1: Create task
task = requests.post(f"{API}/createTask", json={
    "clientKey": KEY,
    "task": {
        "type": "ReCaptchaV3TaskProxyLess",
        "websiteURL": "https://example.com/login",
        "websiteKey": "6LcR_RsTAAAAADHmXXPJh-Lr7Mz4aCezAo3vKWiq",
        "pageAction": "login"
    }
}).json()

task_id = task["taskId"]
print(f"Task created: {task_id}")

# Step 2: Poll for result
while True:
    result = requests.post(f"{API}/getTaskResult", json={
        "clientKey": KEY,
        "taskId": task_id
    }).json()

    if result["status"] == "ready":
        print("Solved!", result["solution"])
        break
    elif result["status"] == "failed":
        print("Failed:", result.get("errorDescription"))
        break

    time.sleep(1)

Using the result

use_result.py

# Step 3: Use the result — paste the token into g-recaptcha-response
token = result["solution"]["gRecaptchaResponse"]
# Browser side: document.querySelector('textarea[name="g-recaptcha-response"]').value = token
# Or include it in your form POST:
resp = requests.post("https://target.example.com/login", data={
    "username": "...",
    "password": "...",
    "g-recaptcha-response": token,
})
print(resp.status_code)

1. Send Payload

Dispatch your ReCaptchaV3TaskProxyLess to our processing cluster via the secure API endpoint.

2. Solving Engine

Capzy's proprietary solver returns valid reCAPTCHA v3 tokens. Final score is determined by Google when your backend calls siteverify — the score depends heavily on the IP that makes the request, so for the highest scores use ReCaptchaV3Task with your own proxy.

3. Get Result

Poll getTaskResult for the validated token, then finalize your automated request.

Using your own proxy

Use ReCaptchaV3Task instead of ReCaptchaV3TaskProxyLess to route the solve through your own proxy — useful when the target site checks that the solving IP matches the submitting IP.

solve_proxy.py

import requests, time

API = "https://api.capzy.ai"
KEY = "capzy_your_key_here"

# Step 1: Create task
task = requests.post(f"{API}/createTask", json={
    "clientKey": KEY,
    "task": {
        "type": "ReCaptchaV3Task",
        "websiteURL": "https://example.com/login",
        "websiteKey": "6LcR_RsTAAAAADHmXXPJh-Lr7Mz4aCezAo3vKWiq",
        "pageAction": "login",
        "proxyType": "http",
        "proxyAddress": "123.45.67.89",
        "proxyPort": "8080",
        "proxyLogin": "user",
        "proxyPassword": "pass"
    }
}).json()

task_id = task["taskId"]
print(f"Task created: {task_id}")

# Step 2: Poll for result
while True:
    result = requests.post(f"{API}/getTaskResult", json={
        "clientKey": KEY,
        "taskId": task_id
    }).json()

    if result["status"] == "ready":
        print("Solved!", result["solution"])
        break
    elif result["status"] == "failed":
        print("Failed:", result.get("errorDescription"))
        break

    time.sleep(1)

Additional proxy parameters

fieldtyperequireddescription

proxyTypetypestringreqyesProxy protocol: http, https, socks4, or socks5

proxyAddresstypestringreqyesProxy IP address or hostname

proxyPorttypenumberreqyesProxy port number

proxyLogintypestringreqnoProxy username (if auth required)

proxyPasswordtypestringreqnoProxy password (if auth required)

userAgenttypestringreqnoUser-Agent string to use. Must match the UA you use when submitting the token

Task Parameters

API Spec

fieldtyperequireddescription

typetypestringreqyesReCaptchaV3TaskProxyLess or ReCaptchaV3Task

websiteURLtypestringreqyesThe URL of the page where reCAPTCHA v3 is loaded

websiteKeytypestringreqyesThe sitekey found in the page source (starts with 6L...)

pageActiontypestringreqnoThe action parameter passed to grecaptcha.execute(). Found in the site's JS code

isEnterprisetypebooleanreqnoSet true if the site uses reCAPTCHA Enterprise

Response Shape

fieldtypedescription

gRecaptchaResponsetypestringThe solved token — submit this in the g-recaptcha-response field or the callback parameter

userAgenttypestringThe User-Agent string the solver used to mint the token. Submit it from the same UA on your backend if Google's siteverify enforces UA binding.

expireTimetypenumberUnix milliseconds when the token expires (issued time + 120s — Google v3 tokens have a 2-minute lifetime).

Example response

{
  "errorId": 0,
  "status": "ready",
  "solution": {
    "gRecaptchaResponse": "03AGdBq25...<long v3 token, 2-minute lifetime>...",
    "userAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
    "expireTime": 1735689720000
  }
}

Error response

{
  "errorId": 1,
  "errorCode": "ERROR_CAPTCHA_UNSOLVABLE",
  "errorDescription": "Solver gave up."
}

Features

AI-powered — no human workers needed

Flat $0.0015 per solve — no hidden tiers

Optional /reportScore feedback loop — share Google's verified score and see your quality on the dashboard

Enterprise support — works with reCAPTCHA Enterprise

Fast — average 8-9 second solve time

pageAction support — matches the site's expected action

99.9% uptime — redundant solver infrastructure

Pricing & Stats

Per 1,000 solves$2.00

Avg solve time~5.0s

Success rate99%+