iHub Apps

Ready-to-use AI apps for your team. Self-hosted. Any model. No prompting expertise needed.

iHub Apps is a full-stack, open-source AI platform that brings generative AI into your secure corporate environment. Your team starts from a personal start page, picks one of the ready-made apps — or chats freely — and gets answers from the model you choose: OpenAI, Anthropic, Google, Mistral, AWS Bedrock, any OpenAI-compatible gateway, or a local model on your own hardware.

Compose emails, translate, summarize, prepare meetings, analyze documents, transcribe recordings, or let a prompt run every morning as a scheduled task. Ground the answers in your own knowledge — files, websites, iFinder — and extend the platform with more apps, skills, models and workflows from the iHub Marketplace, all without writing code. Seamlessly integrate iHub with your existing IntraFind solutions for a unified platform of search, knowledge-based answers and creative AI applications — free and open source.

iHub is developed by IntraFind Software AG — made with ❤️ from our teams in Berlin, Bonn, Munich + Remote

The Software is free-of-use and "AS-IS" without warranty of any kind. — License Details

For enterprise-grade support, custom features, or professional services, contact us at [email protected].

---

🚀 Quick Start — Run in Under 60 Seconds

No Node.js, no Docker, no dependencies. Download the standalone binary and you're running:

⚡ One-Line Installer (Linux/macOS)

The simplest way to install — a single command handles everything:

curl -fsSL https://raw.githubusercontent.com/intrafind/ihub-apps/main/install.sh | sh

OR

Step 1 — Download for your platform

👉 Download the latest release

Platform	File
🐧 Linux	ihub-apps-v*-linux.tar.gz
🍎 macOS	ihub-apps-v*-macos.tar.gz
🪟 Windows	ihub-apps-v*-win.zip

Restricted environment? Use the .base64.txt files: base64 -d ihub-apps-v*-linux.tar.gz.base64.txt > ihub-apps.tar.gz

Step 2 — Extract and run

# Linux / macOS
tar -xzf ihub-apps-v*-linux.tar.gz
cd ihub-apps-v*
./ihub-apps-v*-linux

:: Windows — extract the .zip, then:
ihub-apps-v*-win.bat

Step 3 — Open iHub and connect a model

Open http://localhost:3001 — the binary's port is set in config.env next to the executable (Docker uses port 3000). On a fresh installation the setup wizard opens: sign in with the shipped administrator account (admin / password123), pick a provider — Google Gemini, Anthropic Claude, OpenAI, Mistral AI or a local server such as LM Studio, Jan.ai, Ollama or vLLM — paste its API key and test the connection (a local server needs neither: just continue). Then choose Add your models to pick the models of that provider you want to use.

🎉 Done! iHub auto-configures everything else on first run. No .env file, no database, no manual setup. You can add more providers and models at any time under Admin → Providers and Admin → Models.

🔐 Change the shipped passwords before you open iHub to others. The admin and user demo accounts are meant for the first sign-in; the admin area warns as long as they still use their shipped passwords. See Authentication Quick Start for local accounts, LDAP, OIDC/SSO and proxy authentication.

Other install methods: One-Line Installer · Docker · npm (for developers)

---

👀 A Look Around

🏠 The start page

Every user lands on a personal start page: a greeting, the chat input of the default app so a conversation starts right away, shortcuts to the apps that matter most to them — their favorites first, then the apps you feature — and the chats they can pick up again. Admins choose the default app, the featured apps and even what / opens under Admin → UI Customization → Start Page.

📝 Apps that start with a form

Apps frame a use case: prompt, variables, tools, sources and model. An app can open its chats with a form instead of an empty chat input — users fill in the fields, add their own text or files and press Start. The filled-in prompt goes out as the first message, and the conversation continues like any other chat.

🤖 Any model — and side by side

Users pick the model per chat (when the app allows it), with descriptions and hints from the admin. Compare mode sends one message to two models and streams both answers next to each other.

🎙️ Transcription

Record in the browser or upload an audio or video file: iHub transcribes it into the message — streaming, with self-hosted Voxtral on vLLM, or with Google's hosted Gemini transcription — and the chat model answers the text. Turn meetings into minutes, interviews into summaries, voice notes into emails. Dictation into the input field and read aloud (text-to-speech) are available too.

This is the beginning of the README. Read the rest at the original. Read the full README ↗