Open Minis vs FrankenOCR

Minis — Local-First On-Device AI Agent Minis is a local-first, on-device AI agent that goes beyond chat. All your data stays on your iPhone — nothing is uploaded to a server by default. Running locally means lower latency and stronger privacy, while the agent still takes real action: browsing the web, accessing your health data, managing your schedule, controlling your smart home, and automating complex tasks. Local-First AI Agent Unlike cloud-only chatbots, Minis runs directly on your device and keeps your data there. The agent executes multi-step tasks autonomously, using tools and native iOS frameworks to get things done — not just answer questions, and not just relay everything to the cloud. Multi-Model, Your Choice Bring your own API key. Choose from leading AI providers: • Anthropic Claude • OpenAI GPT • Google Gemini • DeepSeek • Kimi (Moonshot AI) • OpenRouter • Any OpenAI-compatible endpoint Deep iOS Integration The agent accesses Apple frameworks natively: • HealthKit — steps, heart rate, sleep, workouts, SpO₂ • Calendar & Reminders — create events, check availability, manage tasks • HomeKit — control lights, switches, scenes, and smart accessories • Photos — search, organize, and export from your library • Location & Maps — current location, directions, nearby POIs • Vision — OCR text recognition, object detection, barcode scanning • Speech — real-time transcription and text-to-speech • Weather — current conditions, hourly and daily forecasts Built-in Web Browser Navigate pages, take screenshots, extract content, fill forms, and interact with websites — all driven by the AI agent in the background. Skills System Import or create custom AI skills to extend the agent's capabilities. Share skills with others or build your own for specialized workflows. Persistent Memory The agent remembers your preferences, project context, and recurring patterns across sessions — building a personalized assistant that improves over time, all stored locally on your device. Siri Shortcuts Automate AI tasks with Siri and the Shortcuts app. Send prompts with file attachments, wait for results, and chain into complex workflows — all hands-free. Privacy by Design Your API keys and data stay on-device. Fine-grained permission controls let you decide exactly which personal data the agent can access. No account required. No data collected.

Extract text from an image or scanned PDF without uploading the document. FrankenOCR uses the same pure-Rust Unlimited-OCR engine as the FrankenOCR command-line project. Choose a photo, file, camera capture, tall scrolling screenshot, or scanned PDF. The app reads the input in memory and produces structured Markdown on the device. The flagship recognition path uses a downloaded, verified Baidu Unlimited-OCR model for careful document extraction. Tall screenshots are divided with overlap-aware smart cuts, while PDF pages are processed as a document and low-yield pages are called out instead of silently disappearing. Live Camera mode uses Apple's on-device Vision text recognition for speed. It is less accurate than the flagship FrankenOCR model, and the app labels that tradeoff clearly. Recognized words flow from their camera positions into the capture tray as the view updates. The current app ships one flagship model: Unlimited-OCR. It does not claim to include GOT-OCR2, SmolVLM2, OneChart, or a music-recognition model. Images, PDFs, camera frames, and recognized text stay on the device. Network access is used only to download the static flagship model files.