Ethereal Engine - Local LLM

EtherealEngine lets you run powerful AI language models directly on your Android device — completely offline, completely private.

No cloud servers. No subscriptions. No data collection. Just you and your AI, running locally on your phone.

⚡ TRUE ON-DEVICE AI
All AI processing happens on your device using llama.cpp, a high-performance inference engine optimized for mobile.
Your conversations never leave your phone — not even once.

📥 BROWSE & DOWNLOAD MODELS
Explore a curated selection of mobile-optimized AI models from Hugging Face, or search the entire catalog. Pick the model size that fits your device — from lightweight 1B models to powerful 10B parameter models. Choose the right GGUF variant for your available storage and RAM.

💬 FULL-FEATURED CHAT
• Multiple chat conversations with persistent history
• Per-chat model selection and settings
• Attach text files, code, CSV, JSON, and markdown for context
• Markdown rendering with syntax-highlighted code blocks
• Copy any message or code block with one tap
• Thinking mode for step-by-step reasoning

⚙️ FINE-TUNE YOUR EXPERIENCE
• Adjust temperature, context size, and max tokens per chat
• Control CPU thread usage for the best speed/battery balance
• Choose from 10 languages: English, Arabic, Spanish, French, German, Chinese, Japanese, Korean, Portuguese, and Hindi

📦 SMART MODEL MANAGEMENT
• Download progress with background service — downloads continue even if you close the app
• Pause and resume downloads — HTTP range support picks up where you left off
• See storage usage at a glance
• Load and unload models with one tap

🔒 PRIVACY BY DESIGN
• Zero data collection — no analytics, no tracking, no ads
• No account required — start chatting immediately
• No internet needed after downloading a model
• Your chats are stored locally and protected by Android encryption

🎨 DESIGNED FOR MOBILE
• Clean dark interface built with Material Design 3
• Optimized for ARM devices with NEON acceleration
• Smooth streaming text generation
• Edge-to-edge display support

HOW IT WORKS
1. Browse or search for a model in the Explore tab
2. Download the model variant that fits your device
3. Start a new chat — that's it

Your phone becomes your personal AI assistant. Ask questions, brainstorm ideas, write content, analyze files, get help
with code — all without an internet connection and with complete privacy.

RECOMMENDED SPECS
• 6 GB+ RAM for smaller models (1B–3B parameters)
• 8 GB+ RAM for larger models (7B–10B parameters)
• ARM64 device (most modern Android phones)

EtherealEngine is engineered by Ramzy. If you enjoy the app, consider supporting development through the in-app donate link.