> ## Documentation Index
> Fetch the complete documentation index at: https://docs.edge0.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Download & First Run

> The Edge0 app source and the models are obtained separately. Cloning the app repository does not download any model.

## Official sources

| Content | Location |
| - | - |
| App source | [Edge0 GitHub repository](https://github.com/Edge0-AI/Edge0) |
| Edge0 8B | [Hugging Face](https://huggingface.co/Edge0/Edge0-8B-A1B-preview) · [ModelScope](https://www.modelscope.cn/models/Edge0/Edge0-8B-A1B-preview) |
| Edge0 35B | [Hugging Face](https://huggingface.co/Edge0/Edge0-35B-A3B-preview) · [ModelScope](https://www.modelscope.cn/models/Edge0/Edge0-35B-A3B-preview) |

On every platform, installation is currently done by building from source. Follow the platform guide to prepare the build tools.

## Get the app source

After installing Git, run in a terminal:

```bash theme={null}
git clone https://github.com/Edge0-AI/Edge0.git
cd Edge0
```

The repository contains four platform directories: `macos/`, `windows/`, `ios/`, and `android/`. Keep the full repository structure, because some platforms rely on shared files in other directories.

## Prepare the model by platform

| Platform | Model preparation | First run |
| - | - | - |
| [macOS](/macos-guide) | Build and open the app, then download on the **Models** page | Download → **Load model** → chat |
| [Windows](/windows-guide) | Download in the app, which converts to GGUF locally | Download and convert → **Enable** → chat |
| [iOS](/ios-guide) | On a Mac, place the model into `Models/` (35B needs a file-layout conversion first), then build the app | Install → select the model → chat |
| [Android](/android-guide) | Download and convert to GGUF on a computer, then import it to the device via adb | Import → **Refresh** → **Load** → chat |

The model file formats differ by platform and are not interchangeable. Use the files prepared according to the matching platform guide.

## Before you download

1. Confirm your device meets the [Device Requirements](/device-requirements).
2. Decide between 8B and 35B; for a first try, 8B is recommended.
3. Obtain models only from the official sources above.
4. Keep a network connection during the download and the first build.
5. Complete all conversion, packaging, and verification steps in the platform guide.

For model specifications and sizes, see [Models: 8B vs 35B](/models-8b-vs-35b).

## First load

A model must be activated before you can chat. On macOS, click **Load model**. On Windows, conversion starts automatically after the download finishes; then click **Enable**. On iOS, bundle the model into the project before building. On Android, import the model to the device, tap **Refresh**, then tap **Load**.

The first load and the first turn may take longer. Once the model is ready, send a short message to confirm you can get a reply. For an example, see [Quick Start](/quick-start).

If a download, conversion, or load fails, note the error message shown in the interface, then see [Troubleshooting](/troubleshooting).


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