This application provides a local, Flask-based web interface for querying a Llama language model, generating step-by-step reasoning, and visualizing this process as an interactive knowledge graph. It is designed for users who want to explore complex queries and understand the underlying reasoning process locally, without relying on external APIs.

How It Works

The application utilizes a local Llama language model served via Ollama (defaulting to http://localhost:11434) to process user queries. It generates a step-by-step reasoning chain, which is then transformed into a knowledge graph using NetworkX. Semantic similarity, calculated using Annoy and scikit-learn, is employed to find and display related questions and answers.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Run the application: python app.py
• Requires a local Llama language model (e.g., llama3.1:8b) running on http://localhost:11434.
• Prerequisites: Python 3.7+, Flask, NumPy, scikit-learn, Annoy, NetworkX.

Highlighted Details

• Real-time, interactive knowledge graph visualization of the reasoning process.
• Displays the strongest reasoning path.
• Finds related questions and answers based on semantic similarity.
• Local processing via a Llama language model.

Maintenance & Community

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

Licensing & Compatibility

The README does not specify a license. Compatibility for commercial or closed-source use is not stated.

Limitations & Caveats

The application strictly requires a locally running Llama model accessible at http://localhost:11434. No alternative model providers or endpoints are mentioned. The project status (e.g., alpha, beta) and potential for breaking changes are not indicated.