Skip to main content

Installation

This document outlines the detailed procedures for installing Peerz, a network designed to leverage decentralized technologies for the collaborative development and utilization of large language models (LLMs). Aimed at technical users, this guide facilitates both automated and manual installation methods, ensuring you're well-prepared to contribute to or engage with the Peerz network.

Understanding Peerz Components​

Peerz is composed of several pivotal components, each serving a unique function within the ecosystem:

  • Compute Node: Manages AI model computations, essential for the network's distributed computing framework.
  • Validator: Assures the integrity of contributions and updates, maintaining compliance with network protocols.
  • Sequencer: Guarantees the orderly and consistent processing of transactions, critical for network operation.

Familiarity with these components is crucial for navigating the Peerz network effectively.

Pre-installation Considerations​

Prior to installing Peerz, ensure your system meets the following prerequisites. If you're unfamiliar with Peerz and its core concepts, we recommend reviewing the Core Concepts documentation to better understand the network and its ecosystem.

System Requirements​

  • Operating System: Linux (Ubuntu 18.04 or later, CentOS 7 or later) or macOS (10.14 Mojave or later).
  • Python: Version 3.8 or higher installed on your system.
  • Free Disk Space: At least 5 GB of free space to accommodate the installation and future operations.
  • Hardware: A GPU is recommended but not required for Compute Node operations. Ensure you have a CUDA-compatible GPU for optimal performance if intending to utilize GPU capabilities.

Installation Methods​

Automated Installation for Linux and macOS​

Peerz simplifies setup with an automated script, compatible with Linux and macOS, to streamline the environment preparation, including Python virtual environment creation and dependency management.

Run the following command in your terminal to initiate automated installation:

curl -o- https://raw.githubusercontent.com/peerz-ai/peerz/main/scripts/install.sh | bash

The script executes several actions:

  • Verifies Python 3.8 or higher availability.
  • Establishes a Python virtual environment dedicated to Peerz.
  • Installs Peerz and its requisite dependencies within the virtual environment.

Manual Installation Steps​

For users requiring detailed control over the installation or customization according to specific preferences, the manual installation route offers flexibility:

  1. Clone the Peerz Repository: Begin by cloning the Peerz GitHub repository to your system and navigate into the directory:

    git clone https://github.com/peerz-ai/peerz.git
    cd peerz

  2. Create a Python Virtual Environment: Isolating Peerz in a virtual environment is advised to prevent dependency conflicts:

    python3 -m venv .venv
    source .venv/bin/activate

  3. Install Peerz: Within the activated virtual environment, proceed to install Peerz via pip:

    pip install -e .

Confirming the Installation​

After completing the installation, verify the setup by:

  1. Activating the virtual environment if you utilized one during installation.
  2. Executing a test command to confirm Peerz installation: 
    peerz --version
    or for help documentation: 
    peerz -h

Success at this stage indicates Peerz is correctly installed and operational on your system.

Conclusion​

This installation guide is crafted to ensure technical users can efficiently prepare their environments for active participation in the Peerz network. By adhering to the provided automated or manual installation instructions, users can ready their systems for contributing to the decentralized development and application of AI technologies facilitated by Peerz.