Project KAMRAD.

Node for Autonomous Media, Archives, and Data

Knowledge That Never Goes Offline

🌐 English | Русский | Deutsch | Қазақша

---

Project KAMRAD is a self-contained, offline-first knowledge and education server with built-in AI chat, libraries, maps, and tools — all in one place, no internet required.

Features

Feature	Powered By	What You Get
Knowledge Library	Kiwix	Offline Wikipedia, medical references, guides, ebooks
AI Assistant	Ollama + Qdrant	Local chat with document upload and semantic search
Education	Kolibri	Khan Academy courses, progress tracking
Offline Maps	ProtoMaps	Downloadable regional maps with search and navigation
Data Tools	CyberChef	Encryption, encoding, hashing, and data analysis
Notes	FlatNotes	Local note-taking with Markdown support
Benchmark	Built-in	Hardware performance scoring

Quick Install

Requires: Debian-based OS (Ubuntu recommended). sudo/root privileges required.

sudo apt-get update && sudo apt-get install -y curl && curl -fsSL https://raw.githubusercontent.com/Yumash/kamrad/refs/heads/main/install/install_kamrad.sh -o install_kamrad.sh && sudo bash install_kamrad.sh

After installation, open your browser: http://localhost:8080 (or http://DEVICE_IP:8080)

Hardware Requirements

Minimum

• CPU: 2 GHz dual-core
• RAM: 4 GB
• Storage: 5 GB free
• OS: Debian-based (Ubuntu recommended)

Optimal (for AI)

• CPU: AMD Ryzen 7 / Intel Core i7
• RAM: 32 GB
• GPU: NVIDIA RTX 3060 or equivalent (more VRAM = larger models)
• Storage: 250 GB SSD

Multilingual

KAMRAD supports multiple UI and content languages:

• 🇷🇺 Русский (primary)
• 🇬🇧 English
• 🇩🇪 Deutsch
• 🇰🇿 Қазақша

Content collections (Wikipedia, education) are available per language. Maps are global.

Privacy

• Zero telemetry — no data collected or transmitted
• Local-first — everything stays on your device
• No accounts — works without registration
• Internet optional — only needed to download content

License

Apache License 2.0 — see LICENSE

Attribution

Based on Project N.O.M.A.D. by Crosstalk Solutions LLC. Original project licensed under Apache License 2.0.

Contributing

Pull requests and issues are welcome! See CONTRIBUTING.md for details.