How to Set Up Hermes Agent on an Ubuntu Server

Learn how to set up the Hermes AI agent on an Ubuntu server. A step-by-step guide to installing, configuring, and running your own smart assistant.

Dilshad Akhtar
Dilshad Akhtar
Last updated: 8 June 2026
6 min read

TLDRQuick Summary

  • Hermes AI agent is a powerful server-based tool that remembers, browses, codes, and runs files
  • One-command install script handles Python, Node.js, and all dependencies automatically
  • Simple setup wizard configures your agent for daily use
  • Connect to local models via Ollama for privacy or cloud models via OpenRouter for power
  • The hermes gateway command activates your AI assistant on your Ubuntu server

The Agent That Grows With You, and everyone wants it. Let's break down the Hermes AI agent. Instead of relying on external services, it's a powerful tool designed to live directly on your own infrastructure. It retains context and handles real workflows like browsing the web, writing code, and executing files natively.

Minimum system requirements for Hermes AI agent

Why Choose Hermes?

Many people ask about hermes vs openclaw. Both are great tools for building assistants. Openclaw is very fast to start and has a large community. However, Hermes learns better over time. It builds a permanent memory of your projects. If you want a visual app on your personal computer later, you can also use hermes desktop.


OpenClaw and Hermes Agent offer two different approaches to task automation. OpenClaw works as a fast, highly connected tool with a massive library of instant plugins. It lets you set things up quickly and keep tight control over every choice, though it forgets past chats and needs frequent updates to avoid crashing. On the other hand, Hermes Agent acts like an independent worker that learns from its mistakes, writing its own new skills and remembering your context well. While Hermes takes longer to install and connects to fewer apps, it provides better daily stability and safe spaces for running complex code.

Step 1: Install the Agent

Open your Ubuntu server terminal. You need to download and run the install script. Copy and paste this exact code:

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash

This simple script does all the hard work for you. It safely installs Python, Node.js, and other tools you need. You do not have to install them one by one.

Step 2: Configure Your Settings

Once the install finishes, you must set up your agent. Type this code into your terminal:

hermes setup

This will start a short wizard. It will ask you a few simple questions to get your agent ready for daily use. Here is what the wizard covers:

  1. System and Setup — You need a computer running Linux, macOS, or Windows with WSL2. Make sure you have Python and Node installed. Your computer should have at least four gigabytes of RAM. If you want to run models at home, you will need a strong graphics card.

  2. AI Model Details — You must pick an AI provider, like OpenAI or Anthropic. Then, paste in your secret API key. Finally, choose a specific model. The model needs a large memory window so it can handle long tasks without forgetting.

  3. Setting Files — Hermes uses a few files to keep things neat. You put your secret keys in a hidden file. A config file holds your main settings. Two text files, named Soul and Memory, tell the agent how to act and what facts to keep in mind.

  4. Running Commands — Hermes can run terminal commands for you. You must tell it where to run them. It can run right on your machine, safely inside a Docker container, or on a remote server. Docker is the safest choice to protect your system.

  5. Chat Apps and Plugins — You can connect your agent to chat apps like Discord or Telegram. To do this, you need a bot token from those platforms. You can also add special plugins. These give your agent extra tools, like the ability to read your private code.

Step 3: Choose Your AI Brain

Your new agent needs a brain to think. You can connect it to different services.

  • If you want to use models on your own machine for total privacy, you can link it to ollama.

  • If you want to use powerful cloud models, you can easily connect it to openrouter.

To pick your model, run this exact code:

hermes model

Just select the option you like best from the menu that appears.

Hermes agent terminal

Step 4: Start the Gateway

Finally, you need to turn the agent on so it can listen to you. Run this exact code:

hermes gateway

Your AI assistant is now alive on your Ubuntu server! You can start chatting with it and let it help you work faster.

Conclusion

Your AI assistant is now alive on your Ubuntu server! You can start chatting with it and let it help you work faster. You can find in-depth details about the setup process here.

Hermes Agent isn't a chatbot. It's an operating system for personal AI. The memory compounds. The skills compound. The more you use it, the less you have to explain. That's the right design for where this technology is heading.

If you've been waiting for an open-source agent that doesn't require a PhD to set up and doesn't forget your name between sessions, this is it. Install it. Connect Telegram. Give it a project. See what happens after a week.

Then see what happens after a month.

Want to Deploy Your Own Hermes Agent?

Let us handle the setup, configuration, and deployment of your AI agent on a secure Ubuntu server.