openTihui
A native iOS and iPadOS app that runs GGUF language and vision models fully on-device through llama.cpp, or connects to any OpenAI-compatible cloud API, with keyboard shortcuts for quick replies.
openTihui is a SwiftUI application for iPhone and iPad that runs GGUF models locally using llama.cpp's Metal backend on capable Apple GPUs, with a CPU fallback on older devices, or alternatively talks to an OpenAI-compatible cloud endpoint such as OpenAI, OpenRouter, Groq or DeepSeek through the same chat interface. It supports text and image input for vision models, downloads models directly from Hugging Face, and shows streaming Markdown output with a live context-usage meter and a tokens-per-second readout.
Multiple conversations are saved to disk with their own model, system prompt and configuration, and can be exported as PDF or JSON. A shortcuts feature bundles a system prompt, a preferred model and per-chat settings into a reusable preset, such as a translator or a grammar-fixer, which can also be triggered from the iOS keyboard or the Shortcuts app and exported or imported as a JSON file.