LlamaBox: Offline AI Chat

LlamaBox is a fully offline AI chat assistant that runs large language models directly on your Android device. No internet required. No data sent to servers. 100% private.

• Fully Offline — Chat with AI without any internet connection. Your conversations stay on your device.
• Private by Design — No account needed. No cloud API calls. Everything runs locally.
• Download Models — Easily download optimized GGUF models directly within the app.
• Import Models — Bring your own GGUF model files.
• Voice Output — Listen to AI responses with built-in text-to-speech.
• System Monitor — Live RAM, CPU, and GPU usage while chatting.
• Chat History — All conversations saved locally with full history.

LlamaBox uses the llama.cpp engine to run quantized GGUF models on your phone's CPU or GPU. Models like Qwen2.5 0.5B and SmolLM2 360M run smoothly even on mid-range devices.

No data collection. No tracking. No analytics. Your chats are yours alone.