This project provides a reverse-engineered API to access various large language models (LLMs) like Claude and ChatGPT through Quora's Poe platform. It is intended for educational and research purposes, enabling developers and researchers to interact with these powerful models programmatically.

How It Works

The API leverages the requests library to make HTTP calls to Poe's backend services. It requires specific authentication cookies (Quora-Formkey and m-b) obtained from a logged-in Quora session. By mimicking the requests made by the Poe web interface, it allows users to send prompts and receive responses from different underlying LLMs.

Quick Start & Requirements

• Install: pip install requests
• Prerequisites:
  • Quora-Formkey: Obtain from Quora.com page source.
  • Cookie: 'm-b=xxxx': Obtain from Quora.com cookies.
• Example usage: example.py

Highlighted Details

• Access to multiple LLMs: Sage (OpenAI capybara), GPT-4 (OpenAI beaver), Claude+ (Anthropic a2_2), Claude (Anthropic a2), ChatGPT (OpenAI chinchilla), Dragonfly (OpenAI nutria).
• Requires manual cookie extraction for authentication.

Maintenance & Community

No information on contributors, community channels, or roadmap is provided in the README.

Licensing & Compatibility

The README does not specify a license. The project is for educational and research purposes only, and use for other purposes is at the user's own risk.

Limitations & Caveats

This is a reverse-engineered API, meaning it is subject to breaking changes if Quora modifies its internal API. Authentication relies on manually extracted cookies, which may require frequent updates. The project is explicitly for educational and research use, and any other application is not endorsed or supported.