This project provides private and secure AI tools for enhanced productivity, targeting individuals and teams who want to leverage AI for tasks like chatting with AI models and documents. It offers local inference, built-in authentication, user management, and self-hosting optimization for quick deployment.

How It Works

The system utilizes Docker Compose for easy setup and deployment, integrating with Ollama to support over 100 local AI models. It features a chat interface for interacting with both general AI models and specific documents (currently PDFs). For remote AI access, it supports OpenAI and OpenAI-compatible APIs, allowing configuration via environment variables.

Quick Start & Requirements

• Install: curl -sL https://github.com/SecureAI-Tools/SecureAI-Tools/releases/latest/download/set-up.sh | sh followed by docker compose up -d.
• Prerequisites: Docker, Docker Compose. For GPU acceleration on Linux, Nvidia container toolkit is recommended.
• Setup Time: Under 5 minutes.
• Docs: OpenAI/Compatible APIs Guide

Highlighted Details

• Chat with AI models (e.g., ChatGPT, Mistral, Llama2) and documents (PDFs).
• Local inference via Ollama with support for 100+ models.
• Built-in email/password authentication and user management.
• Optimized for self-hosting with Docker Compose.
• M1/M2/M3 Mac inference speed is noted as "really good."

Maintenance & Community

No specific contributors, sponsorships, or community links (Discord/Slack) are mentioned in the README.

Licensing & Compatibility

The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

Currently supports PDFs for document chat; other file types are on the wishlist. GPU acceleration is recommended for performance but not strictly required, with CPU-only mode being significantly slower.