How to Install Hermes Agent on Ubuntu
July 8, 2026 Β· Reading time: 5 minutes
Hermes Agent is an open-source AI agent by Nous Research that runs on your own server. Unlike cloud solutions, you have full control over your data, memory, and configuration. This guide shows you how to get it running on an Ubuntu VPS in under a minute.
What You'll Need
- VPS with Ubuntu 22.04+ β even the cheapest 1 GB RAM one works (we recommend Contabo from β¬4.40)
- LLM API key β OpenAI, Anthropic, DeepSeek, OpenRouter, or any compatible provider
- SSH access to your server
Step 1: Connect to Server
SSH into your VPS:
Step 2: Run Installer
One command installs everything β Python 3.11, Node.js, ripgrep, ffmpeg, and Hermes itself:
git, curl, and xz-utils β usually pre-installed on Ubuntu.Step 3: Reload Shell and Run
Step 4: Configure LLM Provider
Hermes walks you through configuration on first run:
Supported providers: Nous Portal, OpenRouter, OpenAI, Anthropic, Google Gemini, DeepSeek, xAI Grok, and dozens more.
Step 5: Set Up Telegram Bot (Optional)
Running 24/7 as a Service
Minimum Requirements
| πΎ RAM | 1 GB min, 2 GB recommended |
| βοΈ CPU | 1 vCPU minimum |
| πΏ Storage | 10β20 GB SSD |
| β οΈ GPU | Not required β Hermes calls external LLM APIs |
Troubleshooting
hermes: command not found
Reload shell: source ~/.bashrc or exec bash
Missing API key
Run hermes model and select a provider.
Gateway won't start
Check logs: tail -f ~/.hermes/logs/gateway.log
That's it! Your own AI agent is running on your server, with full data and privacy control. Explore more articles β