mobile: a complete offline AI suite for on-device text, image, and vision generation

What it solves

Off Grid provides a complete, private AI suite that runs entirely on-device, eliminating the need for cloud connectivity and preventing data from leaving the user's hardware. It transforms a phone or Mac into a self-contained AI workstation for text, image, and vision tasks.

How it works

The app leverages native hardware acceleration (NPU on Snapdragon, Core ML on iOS) and optimized libraries like llama.cpp and whisper.cpp to run models locally. It supports a wide range of GGUF models for text generation, Stable Diffusion for image generation, and SmolVLM/Qwen3-VL for vision AI. It also includes an on-device RAG system using a bundled MiniLM model for embedding and SQLite for local storage.

Who it’s for

Users who prioritize data privacy, people working in offline environments, and AI enthusiasts who want to run diverse multimodal AI models on mobile devices or Apple Silicon Macs.

Highlights

• Multimodal Capabilities: Combines text generation, image generation, Stable Diffusion, and vision AI in one app.
• On-Device RAG: Local document analysis and knowledge base search using PDFs and text files.
• Tool Calling: Built-in tools for web search, calculator, and device info with automatic tool loops.
• On-Device Voice: Whisper-powered speech-to-text transcription that happens entirely offline.
• Flexible Model Support: Supports Qwen, Llama, and Gemma models, as well as custom .gguf files and remote OpenAI-compatible servers.