> ## 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.

# Device Requirements

> Before you start, confirm that your runtime device, build tools, and available storage meet the following conditions.

**macOS**

* **System**: macOS 14 or later.
* **Chip**: Apple Silicon M3 or later.
* **Build tools**: Xcode Command Line Tools (including the Metal compiler), CMake 3.24+, Rust 1.88, Node.js 22+, Yarn 1.22.

Measured resident memory on a MacBook Air M3:

| Model | Resident memory |
| - | -: |
| Edge0 8B | About 1.7 GB |
| Edge0 35B | About 4 GB |

This is the model's own resident memory, not a whole-machine memory requirement; the system and other apps also use memory.

Model sizes are shown on the **Models** page. The app checks available space before downloading; if space is insufficient, clear some as prompted and try again.

**Windows**

* **System**: Windows 10 or Windows 11, x64.
* **Graphics**: a Vulkan-capable AMD, NVIDIA, or Intel GPU; integrated graphics also work but may be slower.
* **Memory**: 8 GB or more for 8B; 16 GB or more for 35B. Speed drops when memory is tight.
* **Storage**: at least 30 GB of free disk space.
* **Build tools**: Git, PowerShell 7, the Visual Studio 2022 C++ toolchain, CMake 3.21+, the Vulkan SDK, Rust stable, Node.js LTS, plus Python 3.10+ and NumPy.

**iOS**

* **Runtime device**: iPhone, iOS 17 or later.
* **Build computer**: an Apple Silicon Mac with Xcode installed; the project uses Swift tools 6.1.
* **Installation setup**: enable Developer Mode on the iPhone and select your signing Team in Xcode.
* **Model preparation tools**: Python 3.10+, used to download models and convert 35B.

We tested 8B and 35B on an **iPhone 16 Pro (iOS 26.6.2)**.

An app that includes 8B is about **5 GB**, and one that includes 35B is about **20 GB**. The model is installed together with the app; on the Mac, you also need room for model downloads and the build.

**Android**

* **System**: Android 13 or later (API 33+).
* **Architecture**: arm64-v8a; a Snapdragon 8 Elite-class device is recommended.
* **Memory**: 8 GB or more for 8B; 12–16 GB for 35B.
* **Computer storage**: at least 30 GB of free space for model source files and the converted GGUF.
* **Build tools**: JDK 17+, Android SDK 35, NDK r28 (`28.2.13676358`), CMake 3.21+, Ninja, adb, plus Python 3.10+ and NumPy.

Our reference device is a **Lenovo TB322FC (Snapdragon 8 Elite, 16 GB RAM)**. Models are imported onto the device via adb; see the [Android Guide](/android-guide) for the exact steps.

## Test devices at a glance

| Platform | Device and configuration |
| - | - |
| macOS | MacBook Air, Apple M3 |
| Windows | Intel i7-14700K, AMD Radeon RX 9070 GRE, 48 GB DDR5, Windows 11 |
| iOS | iPhone 16 Pro, iOS 26.6.2 |
| Android | Lenovo TB322FC, Snapdragon 8 Elite, 16 GB RAM |

Actual performance on other devices may differ.

## Next step

Once your device meets the requirements, continue to [Download & First Run](/download-and-first-run).


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