Project IRA 🚀

Project IRA is a professional, decoupled AI Orchestrator designed for developers. It moves beyond simple chat interfaces by implementing a tiered architecture (API $\rightarrow$ Service $\rightarrow$ Data) and integrating real-time web browsing and multimodal capabilities.

✨ Key Features

• Multimodal Brain: Support for text and image inputs (Vision-capable).
• Real-time Intelligence: Integrated web searching and headless scraping to bypass the knowledge cutoff.
• Session Management: Persistent chat history stored in a local SQLite database.
• Model Agnostic: Designed to work with any Ollama-compatible model.
• Liquid Glass UI: A cosmic, minimalist frontend built with React and TypeScript.

🛠️ Architecture

The project follows a decoupled service pattern:

• Interface Layer: FastAPI endpoints for seamless communication.
• Service Layer: Business logic handling AI orchestration and tool loops.
• Client/Data Layer: Direct interactions with Ollama API and SQLite.

🚀 Quick Start (Linux Mint / Ubuntu)

1. Backend Setup

# Navigate to project root
cd project_IRA

# Create and activate virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install --upgrade pip
pip install -r requirements.txt

# Start the server
python main.py

2. Frontend Setup

cd ira-ui
npm install
npm run dev

3. Model Requirement

Ensure you have Ollama installed and the desired model pulled:

ollama pull gemma4:31b-cloud # For text/reasoning
ollama pull llava           # For vision support

🛡️ License

Private Repository - For development and research purposes.