A mobile app to chat with local Ollama models from your phone. Run the Python server on your PC, scan the QR code, and start chatting — with streaming responses, vision model support, conversation branching, speech-to-text via Whisper, and full local SQLite storage. Uses Azure DevTunnels for secure access with no port forwarding.

A cinematic 3D web application for AI-powered interactive storytelling. Features immersive space-to-Earth camera animations, real-time narrative generation using local Ollama models, fandom-based story creation (Naruto, One Piece, Douluo Dalu, etc.), and GitHub OAuth integration. Users can create, fork, and share stories with public/private visibility controls. Frontend connects directly to Ollama with devtunnel/VM support, while Express.js backend handles MongoDB persistence and authentication.

A privacy-first AI chat interface for Ollama with voice input (Whisper STT), text-to-speech (Kokoro TTS), vision model support (LLaVA, Qwen-VL), intelligent web search (SearXNG), and reasoning model integration. Built as a deep dive into understanding local LLMs and AI architectures. Runs 100% locally with zero cloud dependencies.

An open-source platform for creating, managing, and interacting with AI assistants. It supports both API-based language models and in-browser WebLLM models for complete privacy.