Features Overview

What Makes Llumen Special

Llumen is designed for users who want privacy and simplicity without sacrificing functionality. Here's what you get:

🚀 Performance First

• Sub-second cold starts: No waiting around
• Real-time token streaming: See responses as they're generated
• Minimal resource usage: Runs on Raspberry Pi and old laptops
• ~17MB binary size: Smaller than most images
• <128MB RAM usage: Leaves resources for other tasks

💬 Three Chat Modes

Normal Chat

Standard AI conversation for general queries and tasks

Web Search

Search the web and get AI-powered answers with sources

Deep Research

Autonomous agents that conduct in-depth research

📎 Rich Media Support

• PDF uploads: Chat about your documents
• LaTeX rendering: Perfect for math and scientific content
• Image generation: Create images directly in chat
• Markdown support: Full formatting capabilities

🎨 Beautiful Interface

Default

Different Background

Different Background

• Mobile-optimized: First-class mobile experience
• Multiple themes: Choose from beautiful pre-made themes
• Pattern overlays: Optional visual flair
• Dark mode: Easy on the eyes

🔌 Universal API Support

Works with any OpenAI-compatible API:

• OpenRouter (recommended)
• OpenAI
• Local models (Ollama, LM Studio, etc.)
• Custom endpoints

🔐 Privacy & Security

• Self-hosted: Your data stays on your device
• No telemetry: Zero tracking or analytics
• No external dependencies: Works completely offline
• MPL 2.0 licensed: Open source and transparent

Comparison with Alternatives

Feature	Llumen	ChatGPT	Open WebUI
Privacy	✅ Self-hosted	❌ Cloud	✅ Self-hosted
Setup Time	⚡ 30 seconds	⚡ Instant	🐌 Hours
Resource Usage	🪶 <128MB	☁️ N/A	💪 >512MB
Web Search	✅ Built-in	✅ Built-in	❌ Manual
Deep Research	✅ Agents	❌ Powerful	❌ No
Mobile UX	✅ Excellent	✅ Good	❌ Poor
Binary Size	🪶 17MB	☁️ N/A	📦 Container

Next Steps

Chat Modes

Learn about the three chat modes

Installation

Get started in 30 seconds