https://github.com/vinyzu/recognizer
🦉Gracefully face reCAPTCHA challenge with ultralytics YOLOv8-seg, CLIPs VIT-B/16 and CLIP-Seg/RD64. Implemented in playwright or an easy-to-use API.
https://github.com/vinyzu/recognizer
ai automation botright bots captcha captcha-bypass captcha-recognition captcha-solver captcha-solving playwright recaptcha recaptcha-solver recaptcha-verification
Last synced: about 1 month ago
JSON representation
🦉Gracefully face reCAPTCHA challenge with ultralytics YOLOv8-seg, CLIPs VIT-B/16 and CLIP-Seg/RD64. Implemented in playwright or an easy-to-use API.
- Host: GitHub
- URL: https://github.com/vinyzu/recognizer
- Owner: Vinyzu
- License: gpl-3.0
- Created: 2023-09-07T11:48:38.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2025-01-13T10:31:14.000Z (5 months ago)
- Last Synced: 2025-05-08T19:11:35.651Z (about 2 months ago)
- Topics: ai, automation, botright, bots, captcha, captcha-bypass, captcha-recognition, captcha-solver, captcha-solving, playwright, recaptcha, recaptcha-solver, recaptcha-verification
- Language: Python
- Homepage:
- Size: 1.18 MB
- Stars: 165
- Watchers: 3
- Forks: 22
- Open Issues: 10
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
🎠reCognizer
#### reCognizer is a free-to-use AI based [reCaptcha](https://developers.google.com/recaptcha) Solver.
Usable with an easy-to-use API, also available for Async and Sync Playwright.
You can pass almost any format into the Challenger, from full-page screenshots, only-captcha images and no-border images to single images in a list.
#### Note: You Should use an undetected browser engine like [Patchright](https://github.com/Kaliiiiiiiiii-Vinyzu/patchright-python) or [Botright](https://github.com/Vinyzu/Botright) to solve the Captchas consistently.
reCaptcha detects normal Playwright easily and you probably wont get any successful solves despite correct recognitions.
---
## Install it from PyPI
```bash
pip install recognizer
```
---
## Examples
### Possible Image Inputs
### Example Solve Video (Good IP & Botright)
https://github.com/Vinyzu/recognizer/assets/50874994/95a713e3-bb46-474b-994f-cb3dacae9279
---
## Basic Usage
```py
# Only for Type-Hints
from typing import TypeVar, Sequence, Union
from pathlib import Path
from os import PathLike
accepted_image_types = TypeVar("accepted_image_types", Path, Union[PathLike[str], str], bytes, Sequence[Path], Sequence[Union[PathLike[str], str]], Sequence[bytes])
# Real Code
from recognizer import Detector
detector = Detector(optimize_click_order=True)
task_type: str = "bicycle"
images: accepted_image_types = "recaptcha_image.png"
area_captcha: bool = False
response, coordinates = detector.detect(task_type, images, area_captcha=area_captcha)
```
---
## Playwright Usage
### Sync Playwright
```py
from playwright.sync_api import sync_playwright, Playwright
from recognizer.agents.playwright import SyncChallenger
def run(playwright: Playwright):
browser = playwright.chromium.launch()
page = browser.new_page()
challenger = SyncChallenger(page, click_timeout=1000)
page.goto("https://recaptcha-demo.appspot.com/recaptcha-v2-checkbox-explicit.php")
challenger.solve_recaptcha()
browser.close()
with sync_playwright() as playwright:
run(playwright)
```
### Async Playwright
```py
import asyncio
from playwright.async_api import async_playwright, Playwright
from recognizer.agents.playwright import AsyncChallenger
async def run(playwright: Playwright):
browser = await playwright.chromium.launch()
page = await browser.new_page()
challenger = AsyncChallenger(page, click_timeout=1000)
await page.goto("https://recaptcha-demo.appspot.com/recaptcha-v2-checkbox-explicit.php")
await challenger.solve_recaptcha()
await browser.close()
async def main():
async with async_playwright() as playwright:
await run(playwright)
asyncio.run(main())
```
---
## Copyright and License
© [Vinyzu](https://github.com/Vinyzu/)
[GNU GPL](https://choosealicense.com/licenses/gpl-3.0/)
(Commercial Usage is allowed, but source, license and copyright has to made available. reCaptcha Challenger does not provide and Liability or Warranty)
---
## Projects/AIs Used
[YOLO11m-seg](https://github.com/ultralytics/ultralytics)
[flavour/CLIP ViT-L/14](https://huggingface.co/flavour/CLIP-ViT-B-16-DataComp.XL-s13B-b90K)
[CIDAS/clipseg](https://huggingface.co/CIDAS/clipseg-rd64-refined)
[]()
## Thanks to
[QIN2DIM](https://github.com/QIN2DIM) (For basic project structure)
---
## Disclaimer
This repository is provided for **educational purposes only**. \
No warranties are provided regarding accuracy, completeness, or suitability for any purpose. **Use at your own risk**—the authors and maintainers assume **no liability** for **any damages**, **legal issues**, or **warranty breaches** resulting from use, modification, or distribution of this code.\
**Any misuse or legal violations are the sole responsibility of the user**.
---
[](https://discordapp.com/users/935224495126487150)
[](https://ko-fi.com/vinyzu)