OGAM: a privacy-focused offline AI suite for mobile and Mac providing text, image, and vision capabilities

What it solves

Off Grid AI provides a complete, fully offline AI suite for mobile devices (Android, iOS) and macOS, ensuring that no data leaves the device. It eliminates the dependency on cloud-based AI services, providing privacy-centric text generation, image creation, and vision analysis on-device.

How it works

The app leverages native hardware acceleration (NPU on Snapdragon, Core ML on iOS) and optimized libraries like llama.cpp, whisper.cpp, and MNN to run models locally. It supports GGUF models for text, Stable Diffusion for images, and Whisper for speech-to-text. It also includes a local RAG system using a bundled MiniLM model and SQLite for document analysis.

Who it’s for

Users who prioritize data privacy and offline access to AI tools, as well as developers who want to run various open-source LLMs and vision models on their personal hardware.

Highlights

• Multimodal Capabilities: Supports text generation (Llama, Qwen, Phi), image generation (Stable Diffusion), and vision AI (SmolVLM, Qwen-VL).
• On-Device RAG: Local knowledge base for PDFs and text documents using embeddings and cosine similarity.
• Tool Calling: Built-in tools for web search, calculator, and knowledge base search with runaway prevention.
• Flexible Model Support: Ability to bring your own .gguf files or connect to remote OpenAI-compatible servers on a local network.
• On-Device Voice: Real-time Whisper speech-to-text and (in Pro version) Kokoro text-to-speech.
• Cross-Platform: Native performance on Android, iOS, and Apple Silicon Macs.