> ## Documentation Index
> Fetch the complete documentation index at: https://docs-2.manicule.dev/docs/llms.txt
> Use this file to discover all available pages before exploring further.
> For every page in one file, fetch https://docs-2.manicule.dev/docs/llms-full.txt (large).
> To search and read these docs as tools, connect the MCP server at https://docs-2.manicule.dev/docs/api/mcp.
> Every agent surface: https://docs-2.manicule.dev/docs/.well-known/llms-manifest.json

# Download & Install Desktop Workspace (/docs/getting-started/installation)

<!-- agent-signals: reading_time_min: 2 · est_tokens: 913 · updated: 2026-11-11 -->
Related: [Import Existing Projects](/docs/getting-started/import-projects.md), [Cloud Studio Quickstart](/docs/getting-started/quickstart.md), [Starter Templates](/docs/getting-started/templates.md)

> Install the native BhaiCode AI Agentic Development Workspace on macOS, Windows, or Linux.

The **BhaiCode Desktop App** is a native development workspace built with **Tauri 2, Rust, and React 19**. Unlike Electron-based editors that bundle Chromium and Node.js runtimes (often consuming 500MB+ idle RAM), BhaiCode uses your operating system's native webview and executes system tasks through a compiled Rust core.

* **Installer size**: \~7–8 MB
* **Idle memory footprint**: \~45 MB RAM
* **Cold launch time**: \< 350 ms

Download the latest release for your operating system:

**[Download BhaiCode Desktop](https://bhaico.de/download)** →

***

## Supported Platforms

| Platform                  | Architecture                  | Package Format       | Minimum OS                              |
| ------------------------- | ----------------------------- | -------------------- | --------------------------------------- |
| **macOS (Apple Silicon)** | `aarch64` (M1 / M2 / M3 / M4) | `.dmg`               | macOS 12 Monterey or later              |
| **macOS (Intel)**         | `x86_64`                      | `.dmg`               | macOS 12 Monterey or later              |
| **Windows**               | `x64`                         | `.msi` / `.exe`      | Windows 10 (Build 19041+) or Windows 11 |
| **Linux**                 | `x86_64` & `arm64`            | `.AppImage` / `.deb` | Ubuntu 22.04+, Debian 12+, Fedora 38+   |

***

## macOS Installation

1. Download `BhaiCode-darwin-arm64.dmg` (Apple Silicon) or `BhaiCode-darwin-x64.dmg` (Intel).
2. Open the downloaded `.dmg` file.
3. Drag the **BhaiCode** icon into your **Applications** folder.
4. Launch BhaiCode from Spotlight (`Cmd + Space` and type "BhaiCode") or Launchpad.

> **macOS Gatekeeper**: On first launch, macOS may display a standard security prompt. Select **Open** to confirm. BhaiCode binaries are signed and notarized with Apple Developer certificates.

***

## Windows Installation

1. Download `BhaiCode-Setup-x64.exe` or `BhaiCode-x64.msi`.
2. Double-click the installer to start setup.
3. Select your installation directory (defaults to `%LOCALAPPDATA%\Programs\BhaiCode`).
4. Click **Install**. Setup completes in a few seconds.
5. Launch BhaiCode from the Start Menu or desktop shortcut.

***

## Linux Installation

BhaiCode provides both standalone **AppImage** and Debian/Ubuntu **`.deb`** packages:

### AppImage (Universal Linux)

```bash
chmod +x BhaiCode-linux-x64.AppImage
./BhaiCode-linux-x64.AppImage
```

### Debian / Ubuntu (`.deb`)

```bash
sudo dpkg -i BhaiCode-linux-x64.deb
sudo apt-get install -f
```

***

## First-Launch Configuration

When you first open BhaiCode Desktop:

1. **Sign In or Continue as Guest**: Sign in to sync projects across devices, or use **Local-First / BYOK Mode** without an account.
2. **Configure AI Models**: Enter your API key for OpenAI, Anthropic, Google Gemini, Nous Portal, or connect to local offline models running via **Ollama** (`http://localhost:11434`). Keys are stored encrypted in your operating system's native keychain (macOS Keychain, Windows Credential Manager, Linux Secret Service).
3. **Open a Project**: Click **Open Folder** (`Cmd + O` / `Ctrl + O`) to select any local Git repository. BhaiCode inspects the project structure, language dependencies, and Git status immediately.

***

## Next Steps

* [Desktop Workspace Overview](/docs/desktop/overview): Dual-surface architecture and performance details.
* [Agentic Canvas Mode](/docs/desktop/agentic-mode): Delegating multi-file tasks to the agent swarm.
* [Code Editor & Ask AI](/docs/desktop/code-editor): CodeMirror 6 editor and highlight-to-ask actions.