Home / Servers / Hermes Agent VPS
Self-host Hermes Agent on your own VPS
Hermes Agent is a self-improving AI agent built by Nous Research that learns from every session. Running it 24/7 on a VPS means it can keep working, keep learning, and stay reachable from Telegram, Discord, Slack, or your CLI without your laptop being on. Servury is built for this: anonymous signup, crypto payments, full root, no logs, no babysitting.
Hermes runs comfortably on a 2 GB RAM plan. Starting at $15.59/mo.
Why a VPS for Hermes
Quick start: install Hermes on Servury
Tested on Ubuntu 24 and Debian 12. Pick a 2 GB+ plan, deploy, SSH in, paste this.
# 1. System prep
apt update && apt install -y python3-pip python3-venv git tmux
# 2. Install Hermes Agent
python3 -m venv ~/hermes && source ~/hermes/bin/activate
pip install hermes-agent
# 3. Configure your LLM provider (OpenAI / Anthropic / OpenRouter / your own)
hermes init
# 4. Run it persistently in tmux so it survives SSH disconnects
tmux new -s hermes
hermes serve
# Ctrl+B then D to detach, the agent keeps running
Prefer Docker? docker run -d --restart unless-stopped --name hermes -v $HOME/.hermes:/root/.hermes ghcr.io/nousresearch/hermes-agent:latest serve works the same.
Recommended specs
Personal assistant on Telegram, occasional CLI use, small skill library.
VPS-100 plan
Multi-channel (Discord + Slack + email), growing skill set, fast response times.
VPS-150 plan
Local LLM endpoint, large knowledge base, multiple concurrent users.
VPS-200 plan or VDS
Frequently asked questions
What is Hermes Agent?
Hermes Agent is an open-source self-improving AI agent from Nous Research. Unlike a chatbot wrapper, it persists across sessions, learns skills from experience, and connects to multiple messaging platforms (Telegram, Discord, Slack, email, CLI). Running it on a VPS means it stays online and keeps learning even when your laptop is closed.
How much VPS do I need to run Hermes?
2 GB RAM and 10 GB disk is the official minimum. For real day-to-day use we recommend 4 GB RAM. The agent itself is light, but if you wire in a local LLM you will need more.
Does Servury log what my Hermes agent does?
No. Servury runs no application-level logging. We do not have access to anything inside your VM. With anonymous registration and crypto payment, even the billing record contains no personal info.
Which OS should I pick?
Ubuntu 24 or Debian 12. Both work great for the official Hermes Agent install instructions and for Docker-based deployment.
Can I run a local LLM on the same VPS instead of paying OpenAI/Anthropic?
Possible but not recommended on small plans. Local LLMs need 16+ GB RAM and ideally a GPU. If you want full local inference, look at a VDS plan or check our owned hardware in Montreal which has more headroom.
Will my agent stay running after I close my SSH session?
Yes, if you use tmux, screen, systemd, or Docker. We show the tmux pattern in the quick-start above. For production use, a systemd service or Docker with --restart unless-stopped is cleaner.
Can I pay with crypto?
Yes. XMR, BTC, LTC, ETH, USDT, USDC and more via our crypto processor. No KYC, no email confirmation, no name on file.
Where are servers located?
Montreal (owned hardware), New York, London, Paris, Frankfurt, Netherlands, and Singapore. Pick whichever is closest to where you usually message your agent from for lower latency.
