> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zerogpu.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Register as an Edge Provider, connect your first device with the ZeroGPU Edge agent or an SDK, and start earning.

This guide takes you from a new account to a device serving inference on the ZeroGPU edge network.

**1. Register your app or fleet**

Sign up at the [Edge Operator Portal](https://edge.zerogpu.ai/auth/sign-up), then choose **Add New App** and pick what you're connecting:

> [tab] App
>
> For an app your users run: Telegram Mini App, Chrome extension, website, or Android.
>
> | Field | What to enter |
> | - | - |
> | **App name** | The name of your app |
> | **Website / store URL** | Your website, store listing, or Telegram `t.me` link |
> | **Category** | Productivity, Utilities, Creative, Finance, News, Lifestyle, or Gaming |
> | **Estimated users** | Your approximate monthly active users |
> | **Primary region** | Where most of your users are |
> | **App types** | The platforms you'll integrate: Telegram Mini App, Chrome Extension, Browser, Android. For a macOS desktop (Electron) app, choose **Others** and describe it. |

> [tab] GPU Fleet
>
> For machines you manage with the downloadable agent (PCs, Macs, lab or office computers, servers, cloud VMs), and for servers or containers running `@zerogpu/worker`.
>
> | Field | What to enter |
> | - | - |
> | **Fleet name** | A name for this group of machines |
> | **Website / organization URL** | Your organization's website |
> | **Category** | University / Research Lab, Gaming Center, Cloud / Data Center, Enterprise IT, or Other |
> | **Estimated devices** | How many machines you plan to connect |
> | **Primary region** | Where most of the machines are |
> | **Device types** | macOS, Windows, and/or Linux / Server |

Accept the terms and select **Submit for approval**. Your app or fleet shows as **Pending Review** on **My Apps**; review usually takes 24–48 hours. Once it's **Approved**, open it in the portal to copy its **SDK key**.

> [!NOTE]
>
> The SDK key (`zgpu-sdk-…`) is for Edge Providers: it lets your devices join the network, and ZeroGPU Edge calls it the *operator key*. It is different from the API key (`zgpu-api-…`) used to *call* the ZeroGPU API. Don't mix them up.

**2. Select your platform**

**Downloadable agent**, for machines you manage. Get it from [apps.zerogpu.ai/edge](https://apps.zerogpu.ai/edge).

> [card] [Windows](https://docs.zerogpu.ai/edge/agent/windows.md)

> [card] [macOS](https://docs.zerogpu.ai/edge/agent/macos.md)

> [card] [Linux](https://docs.zerogpu.ai/edge/agent/linux.md)

**SDK integration**, for apps you ship.

> [card] [Android](https://docs.zerogpu.ai/edge/sdk/android.md)

> [card] [Chrome Extension](https://docs.zerogpu.ai/edge/sdk/chrome-extension.md)

> [card] [Browser](https://docs.zerogpu.ai/edge/sdk/browser.md)

> [card] [Telegram Mini Apps](https://docs.zerogpu.ai/edge/sdk/telegram.md)

> [card] [macOS desktop apps](https://docs.zerogpu.ai/edge/sdk/electron.md)

> [card] [Server / npm](https://docs.zerogpu.ai/edge/sdk/server.md)

Each guide covers requirements, downloads, setup, and troubleshooting. [Platforms and downloads](https://docs.zerogpu.ai/edge/platforms.md) compares them all.

**3. Install**

> [tab] Android
>
> ```kotlin build.gradle.kts theme={null}
> dependencies {
>     implementation("ai.zerogpu:android-sdk:0.5.0")
> }
> ```

> [tab] Web
>
> ```html theme={null}
> <script async src="https://js-sdk.zerogpu.ai/dist/loader.js"></script>
> ```

> [tab] Chrome extension
>
> ```bash theme={null}
> npm install @zerogpu/browser-extension-sdk
> npx zerogpu init
> ```

> [tab] Telegram Mini Apps
>
> ```html theme={null}
> <script async src="https://js-sdk.zerogpu.ai/dist/loader.js"></script>
> ```

> [tab] macOS desktop app
>
> ```bash theme={null}
> npm install @zerogpu/worker
> ```

> [tab] Server
>
> ```bash theme={null}
> docker pull zerogpu/worker
> # or
> npm install -g @zerogpu/worker
> ```

> [tab] Agent
>
> Download the installer for each machine from [apps.zerogpu.ai/edge](https://apps.zerogpu.ai/edge):
>
> | OS | Guided install | Managed / silent install |
> | - | - | - |
> | Windows 10/11 | `ZeroGPU-Setup.exe` | `ZeroGPU-Edge.msi` |
> | macOS 14+ (Apple silicon) or 13.4+ (Intel) | `ZeroGPU.dmg` | `zerogpu-edge.pkg` |
> | Linux | `.deb` (Ubuntu, Debian) or `.rpm` (Fedora, RHEL, Rocky) | the same packages |

**4. Add your SDK key**

Keep the key out of source control and inject it at build or deploy time.

> [tab] Android
>
> ```kotlin theme={null}
> val sdk = ZeroGpuSdk(
>     context = applicationContext,
>     operatorKey = BuildConfig.ZEROGPU_OPERATOR_KEY, // "zgpu-sdk-…"
> )
> ```

> [tab] Web / Telegram
>
> ```html theme={null}
> <script
>   async
>   src="https://js-sdk.zerogpu.ai/dist/loader.js"
>   edgeOperatorKey="zgpu-sdk-your-key"
> ></script>
> ```

> [tab] Chrome extension
>
> ```js offscreen.js theme={null}
> import { initZeroGpuOffscreen } from "./vendor/zerogpu/zerogpu-offscreen.js";
>
> initZeroGpuOffscreen({ edgeOperatorKey: "zgpu-sdk-your-key" });
> ```

> [tab] macOS desktop app
>
> ```js theme={null}
> const worker = new ZeroGpuWorker({ operatorKey: ZEROGPU_OPERATOR_KEY });
> ```

> [tab] Server
>
> ```bash theme={null}
> docker run -d -e ZG_OPERATOR_KEY_FILE=/run/secrets/zg_key \
>   -v /path/to/zg_key:/run/secrets/zg_key:ro \
>   -v zg-models:/models -v zg-state:/data zerogpu/worker
> ```

> [tab] Agent
>
> The guided installers ask for the key. For managed installs, pass it on the command line:
>
> ```bash theme={null}
> # Windows
> msiexec /i ZeroGPU-Edge.msi /qn OPERATOR_KEY="zgpu-sdk-your-key"
>
> # macOS and Linux, after installing the package
> sudo zerogpu-edge configure --operator-key "zgpu-sdk-your-key" --restart
> ```

**5. Start contributing compute**

Registration is automatic: with a valid key, the device joins the network by itself.

* **Android and macOS desktop apps:** call `start()`.
* **Browsers, Telegram Mini Apps, and Chrome extensions:** the SDK starts as soon as it's initialized with your key.
* **Server:** the worker starts with the container or process.
* **Agent:** ZeroGPU Edge starts as a background service and starts again at every boot.

From then on the device serves inference tasks from the network. You never feed it tasks or poll it.

**6. Confirm the device is online and earning**

Check **My Apps** and **Analytics** in the [Edge Operator Portal](https://edge.zerogpu.ai). Apps and fleets serving traffic show as **Active/Earning**. [Verify devices](https://docs.zerogpu.ai/edge/verify.md) covers every platform's checks, and [Earnings and payouts](https://docs.zerogpu.ai/edge/earnings.md) explains what you're paid.

A connected device only receives work when the network needs it, so it may sit ready for a while before its first task.

> [!WARNING]
>
> **Tell your users.** In an app, show your users the End User Disclosure from the Edge Operator Portal, word for word, with your app's name, before the SDK starts, and get their agreement. Give them a way to turn resource sharing off.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.