#560 · Primary category: AI Agents & Automation

hcaptcha-challenger

agent ai-agents captcha captcha-solver captcha-solving chatgpt clip gemini hcaptcha hcaptcha-solver llm openai playwright yolo

🥂 Gracefully face hCaptcha challenge with multimodal large language model.

Project last updated:08/15/26

GitHub Stars

2.5K

Forks

446

Contributors

12

License

GPL-3.0

Why we included this project

Browser automation hits a hard stop the moment a site decides to throw an hCaptcha at it. This Python library gives Playwright-based agents a way through by solving the common challenge types in-process. Small ONNX vision models handle the classification and detection work (ResNet, YOLOv8, ViT), and a multimodal LLM steps in for the trickier drag-drop and agentic tasks. It needs no Tampermonkey script and no third-party anti-captcha service, so the whole flow stays under your control. For teams running sign-in, checkout, or scraping pipelines that have to survive CAPTCHA gates without a human hovering nearby, the bundled model pack and pluggable per-challenge resources are a solid place to start testing against real traffic.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category