Hermes Agent Setup on VPS

Summary

You'll learn how to set up a Hermes AI agent on a VPS (Virtual Private Server) in just four simple steps. This method is recommended because it provides 24/7 runtime, keeps your agent separate from your local machine, and offers the cleanest installation experience. You'll start by choosing a VPS provider—the video recommends Zebra VPS for its affordability, with plans starting at $2-3 per month for 2-4GB RAM, which is sufficient for an executive assistant role. You'll need your VPS IP address, SSH username, and password to connect. The guide recommends using Termius as your SSH client because it includes SFTP file access, giving you full visibility into your server files. Once connected, you'll install Hermes by copying a single command from the official documentation. If the installation fails on a fresh VPS, you may need to run 'sudo apt update' first to install Git and curl dependencies. During setup, you'll choose between quick setup (recommended for beginners) or full setup for more control. You'll select your AI model provider—the video uses MiniMax M2.7, but you can choose from Anthropic, DeepSeek, Gemini, or use an OpenRouter API key. After entering your API key and selecting your model, you can optionally connect a messaging platform like Discord. The entire process takes just minutes, and once complete, your Hermes agent runs continuously on the VPS, ready to assist you 24/7 with research and tasks without consuming your local machine resources.