Creating your own HomeLAB using Antigravity: Quad Architecture (v3.2)

### Introduction
Welcome to **HomeLAB v3.2**, the "Sovereign Stack" AI Orchestrator. This setup gives you a multi-layered AI system that combines high-end cloud intelligence (Gemini/Kimi) with local privacy (Gemma 3).

### Prerequisites
1. **Abacus.ai Pro Package**: Sign up for the $20/month Pro package at [Abacus.ai](https://abacus.ai). This provides access to the powerful **Gemini 3 Flash** and **Kimi k2.5** models used in the Strategy and Review departments.
2. **Antigravity AI**: This project is designed to be set up and managed by **Antigravity**. Download it here: [https://antigravity.google/download](https://antigravity.google/download).

### Installation Steps
1. **Create a Folder**: Create a folder on your computer (e.g., on your `D:` or `C:` drive) exactly called `VAULT-AI`.
2. **Download & Unzip**: Download the HomeLAB Zip and move it into your new `VAULT-AI` folder. Unzip it there.
3. **Launch Antigravity**: Open Antigravity, go to **File** ➔ **Add Folder to Workspace**, and select the `VAULT-AI` folder you just created.
4. **Automated Setup**: Simply ask Antigravity:
> "Set up my HomeLAB"

Antigravity will automatically handle the virtual environment, install dependencies, and walk you through adding your API keys.

### The Triad Architecture (LLM Fallback System)
HomeLAB v3.2 uses a resilient "N+1" redundancy system. If a primary service is down, it automatically switches to a fallback model:
- **Strategy Dept**: `Gemini 3 Flash` (Primary) ➔ `Gemini 2.5 Flash` (Fallback).
- **Antigravity Dept**: `Gemini 3.1 Pro` (Primary) ➔ `Gemini 3 Pro` (Fallback).
- **Review Dept**: `Kimi k2.5` (Primary) ➔ `Kimi k2.0` (Fallback).
- **Local Dept**: `Gemma 3:27b` (Primary) ➔ `Gemma 3:12b` (Fallback).

### Smart Memory & Safety
- **Context Memory**: Automatically summarizes conversation history every 10 exchanges, allowing for massive "infinite" logic retention without crashing the context window.
- **Auto-Backup**: The system automatically zips its own core engine files into the `/backups` folder periodically, ensuring your custom logic is never lost.
- **VRAM Management**: High-speed "Purge" logic for local models, optimized for AMD RX 7900 XTX (ROCm) and NVIDIA alike.

---
*Note: I've shared this to help everyone build their own sovereign AI workstation. My only secret is my Abacus key – so make sure to get your own! Enjoy the Triad!*

If for some reason you do not want to sign up with ABACUS.AI -- that's OK, you can still use the Local Dept: (Gemma 3: 27b) or (Gemma 3:12b) models completely FREE of charge -- No API_KEY is required to run the Local models
Enjoy!
Best
Xard777