Open MobileUI vs HAIchat

Open MobileUI is an open-source, native iPhone client for Open WebUI. Connect to your self-hosted server and chat with your AI models on the go: local Ollama models, OpenAI-compatible APIs, or any LLM your Open WebUI admin has set up. Your server, your data. The app connects directly to the Open WebUI server you choose. We do not run any backend that receives, stores or processes your chats, and there are no ads. CONNECT THE WAY YOU HOST • Home lab, VPS or company server • Private networks and Tailscale, including plain HTTP addresses for local setups • Custom domains behind a reverse proxy such as Nginx, Caddy or Traefik • Encrypted HTTPS and WSS connections • Sign in with email and password, LDAP, SSO (OpenID Connect), Google, Microsoft or GitHub CHAT • Real-time streaming responses • Edit and regenerate responses • Switch between AI models in one tap • Attach photos, camera shots, files, web pages and knowledge bases • Web search, image generation and code interpreter when your server enables them • Rich rendering: Markdown, code highlighting, math formulas and source citations • Voice mode: talk to your assistant hands-free and hear the answers read aloud • Voice dictation for your messages • Temporary chats that are not saved ORGANIZE • Folders, pinned chats, search and archive • Share chats and export them BUILT FOR IPHONE • Native performance, smooth animations and gestures • Light, dark and system themes • Smart keyboard handling • Graceful handling of network interruptions • Available in English, German, Spanish, French, Portuguese, Russian, Chinese, Japanese and Romanian PERFECT FOR • Using your self-hosted AI assistant on the go • Privacy-conscious users who want full control over their data • Teams that need mobile access to their company AI assistant REQUIREMENTS Open MobileUI does not host models or provide an AI service. It connects to an Open WebUI server that you or your organization run. Available features depend on your server settings. OPEN SOURCE Open MobileUI is free and open source under the GPL-3.0 license. For support, issues and source code: https://github.com/RonasIT/open-webui-react-native Disclaimer: Open MobileUI is an independent, third-party application. It is not affiliated with the Open WebUI, Ollama or Tailscale projects.

HAI Chat is a fast, private client for the Hetzner Inference API, the experimental platform serving open-weight AI models. Bring your own API key, pick a model, and chat. WATCH IT THINK The models reason before they answer, and HAI Chat streams that thinking live in a collapsible fold. You choose how much thinking a question deserves, from off to thorough, per chat or as a default. OPEN MODELS, HUGE CONTEXT Choose between DeepSeek, GLM, Kimi, and Qwen, with context windows up to 512k tokens. The model list loads live from the API, so new models appear as Hetzner adds them. ON IPHONE, IPAD, AND MAC One app for all three. On iPad, on a Mac, and on a larger iPhone held sideways, your chats sit in a sidebar next to the open conversation. With a keyboard attached, Return sends and ⌘N starts a new chat. REPLIES THAT READ PROPERLY Code blocks, tables, headings, lists, and quotes are laid out rather than arriving as raw symbols. When a model answers with a mermaid diagram, you get the drawing, rendered on your device with no network involved. ASK ABOUT PHOTOS Attach up to four pictures and ask. Take them with the camera, pick them from your library, paste them, or drag them in. Models that can see images describe, read, and explain what's in them. KEEP WHAT MATTERS Search every chat by title or by what was actually written. Rename them. Save a single reply as Markdown or as a laid-out PDF, diagrams and all, or export a whole conversation. LONG CHATS KEEP GOING Every chat shows how much of the model's memory it is using and what is taking up the room. When it gets tight, HAI Chat offers to summarise the older messages so the conversation can carry on. Your messages stay exactly as you wrote them. Only what gets sent changes. YOUR TONE, YOUR RULES Pick a response style like Concise or Professional, or write your own. Add a profile once and every chat knows who it's talking to. You can see exactly what gets sent, word for word. TALK AND LISTEN Speak instead of typing, with dictation that runs on your device. Replies can be read aloud while they still stream, in the language they're written in, without reading markdown symbols out loud. Pick a favourite voice per language, listen to samples first, and use your Personal Voice if you have one. SHARE STRAIGHT IN Send text, a link, or a picture to HAI Chat from any app's share sheet. It opens as a new chat with everything already in the message field. Nothing is sent until you tap send. PRIVATE BY DESIGN There is no account and no tracking. Your API key stays in the Keychain and goes only to the Hetzner Inference API. Chats are stored on your device, and nowhere else unless you switch on iCloud sync, which puts them in your own private iCloud database that the developer cannot read. Lock the app behind Face ID if you like. IN ENGLISH AND GERMAN The interface follows your system language. HONEST BY DESIGN AI answers can be wrong. HAI Chat says so right under the message field, and its terms don't pretend otherwise. REQUIREMENTS You need your own API key for Hetzner's experimental Inference platform, free of charge while the platform is in its experimental phase. The platform and its model lineup may change or be discontinued at any time. That is part of the experiment. HAI Chat is an independent project. It is not affiliated with, endorsed by, or operated by Hetzner. "Hetzner" is a trademark of Hetzner Online GmbH, used here only to describe compatibility.