Update README.md

README.md

@@ -1,3 +1,40 @@
----
-license: apache-2.0
----
+# 🧠 OfflineAI 1.01 – NemoMix 12B (macOS Edition)
+
+This repository contains the pre-packaged `.gguf` model and launch script for **OfflineAI 1.01**, an educational, fully offline AI assistant designed to run on macOS without any internet access.
+
+## 💡 About the Project
+
+**OfflineAI.Online** is a community project by David Káninský, created to demonstrate the capabilities of modern open-source language models in an offline and privacy-focused environment.
+
+This version uses the `NemoMix-Unleashed-12B` model (quantized to `Q5_K_M`) and integrates with `llama.cpp`, allowing it to run on machines with ~16 GB RAM.
+
+## 📁 What’s included
+
+- `OfflineAI-Nemo12B-Q5_K_M.gguf` – The model (quantized)
+- `Offline_AI.command` – A one-click launcher script for macOS Terminal
+- `info.txt` – Legal and usage notice
+
+> Simply unzip the folder, double-click `Offline_AI.command`, and chat offline.
+
+## 🧪 Requirements
+
+- macOS (tested on Ventura & Sonoma)
+- ~16 GB RAM minimum
+- No internet connection required after download
+
+## ⚠️ Legal Notice
+
+This model is intended for **educational and demonstrational use only**.  
+It is based on the open-source NemoMix 12B model (MIT license) and distributed in full respect of open-source licensing.
+
+OfflineAI does **not** provide legal, financial, or medical advice. Use outside the scope of demonstration is at your own risk.
+
+## 🙏 Credits
+
+- Based on [NemoMix-Unleashed-12B](https://huggingface.co/bartowski/NemoMix-Unleashed-12B-GGUF) by [@bartowski](https://huggingface.co/bartowski)
+- Powered by [llama.cpp](https://github.com/ggerganov/llama.cpp)
+- Offline integration and launcher: [OfflineAI.Online](https://offlineai.online)
+
+## 💬 Support or feedback?
+
+You can find updates and more details at [OfflineAI.Online](https://offlineai.online)