AI Workstation
Your AI, right on your desktop.
AI Workstation is a native app for macOS, Windows, and Linux that wraps Hrida.ai in a desktop shell. Run models locally with the built-in llama.cpp engine, connect to any Hrida.ai server, or do both at once — no Docker, no terminal, no setup. Download, launch, chat.
On top of the full Hrida.ai web experience, the desktop app adds things only a native app can do: a Spotlight overlay you can summon over any window, system-wide voice input, a local inference engine, a bundled server, and a bundled Hrida Terminal — all managed from one window.
How it works
| 🖥️ Run locally | The app can install and run the Hrida.ai server (Studio) on your machine, and optionally the built-in llama.cpp engine to download and run models fully offline. Nothing leaves your computer. |
| ☁️ Connect remotely | Point the app at any Hrida.ai server. Switch between multiple connections from the sidebar. |
| 🔀 Both at once | A local server and several remote connections can be open side by side. |
Feature map
| Area | What's there |
|---|---|
| Installation | Platform downloads, system requirements, first launch, privacy posture. |
| Connections | Multiple simultaneous servers, local vs remote, default connection, embedded session isolation, in-app links to HridaOne / CRM. |
| Local server | Bundled Hrida AI Studio: one-click install (Python + uv + wheel), start / stop / restart, port, serve on the local network, auto-update, data location, logs, uninstall. |
| Local inference | Built-in llama.cpp engine (CPU / CUDA / Vulkan / ROCm / Metal), version and build management, plus Hugging Face model search, download, and cache management. |
| Spotlight | Floating chat bar over any app, drag-to-screenshot region capture, clipboard auto-paste. |
| Voice & Call | System-wide push-to-talk voice input with automatic transcription, and a global shortcut to start a voice call. |
| Hrida Terminal | Bundled Hrida Terminal server: enable, port, working directory, generated API key, start on launch. |
| Updates | Background auto-updates, the update proxy, and the About panel. |
| Settings & configuration | Language, appearance, launch at login, run in background, global shortcut, environment variables, install location, tray menu, reset, config file locations. |
Download
Grab the installer for your platform from the releases page. Internet is required on first launch; after that everything works offline. See Installation for the full platform matrix and system requirements.
Privacy
No telemetry. No tracking. No phone-home. Conversations stay on your machine. When you run models locally, nothing leaves the computer at all.
License
See the project LICENSE. The bundled Hrida AI Studio and Hrida Terminal wheels are private/commercial components shipped inside the app.