VPS 1
- 1
- vCPU
- 1 GB
- RAM
- 20 GB
- NVMe
From $14.99/mo
Billed monthly- Dedicated IPv4 address
- Full root access
- 99.99% uptime guarantee
- 24/7 expert support
- OS: AlmaLinux, CentOS Stream, Debian
Servers available:Germany Finland Singapore
Hermes Agent Hosting ยท Runs on your own VPS
Hermes Agent hosting for an agent that remembers
Run Hermes Agent from Nous Research on your own VPS with full root access. Its memory and skills stay on your server and the agent keeps working when your computer is off.
- Persistent Memory
- Full Root Access
- Your Own Data
- 24/7 Human Support
- 30-day money-back guarantee
- Full root access
- Cancel anytime
Hermes Agent
Running on your VPS
$ ssh root@your-server-ip
$ curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
$ hermes model
Gateway runningReplies in Telegram
Remembers context
Root access
What our customers sayExcellent4.8 average from 53 reviews on Trustpilot, Google and HostAdvice
What is Hermes Agent
Hermes Agent is an agent that learns as it works
Hermes Agent is an open-source agent from Nous Research under the MIT licence. It keeps memory between sessions, can write reusable skills from tasks it has finished, and can reach you in apps such as Telegram, Discord, Slack, WhatsApp, Signal and email.
Hermes Agent
Gateway running
- TelegramConnected
- DiscordConnected
- MemorySaved on your server
hermes gateway setup
Why a VPS
Give Hermes Agent a home that stays online
Hermes Agent builds up memory and skills over time, and its messaging gateway has to be online to reply. On a VPS the agent keeps running when your own computer is off, and what it has learned stays in one place that you control.
Get Hermes Agent running in 3 steps
You do the install yourself, so you stay in control. Our support team helps if you get stuck.
Order a VPS
Pick a plan and a Linux image such as Ubuntu or Debian. When the server is ready, log in over SSH as root.
ssh root@your-server-ipInstall Hermes Agent
Run the official one-line installer. It sets up what the agent needs, and Git is the only thing you need first.
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bashPick a model and connect a chat app
Choose a model provider, then set up the messaging gateway for the app you want to talk to the agent from.
hermes model
Which VPS size for Hermes Agent
The Hermes docs do not publish a minimum server size. When the model runs at a provider, the agent itself is light, so start with a small plan and move up if it feels slow.
- The model needs a context window of at least 64,000 tokens.
- Local models need far more memory and usually a GPU. Our VPS plans are CPU servers, so use a provider API.
- You pay your model provider directly. WaseerHost only charges for the server.
VPS Hosting Plans
NVMe VPS servers with full root access. Choose your server location and operating system when you order.
VPS 3
- 2
- vCPU
- 2 GB
- RAM
- 40 GB
- NVMe
$34.99/mo
Billed monthlyVPS 5
- 3
- vCPU
- 4 GB
- RAM
- 80 GB
- NVMe
$62.99/mo
Billed monthlyVPS 7
- 8
- vCPU
- 16 GB
- RAM
- 320 GB
- NVMe
From $138.99/mo
Billed monthlyVPS 9
- 16
- vCPU
- 32 GB
- RAM
- 640 GB
- NVMe
From $257.99/mo
Billed monthlyQuestions about Hermes Agent hosting and running Hermes Agent
Hermes Agent is an open-source AI agent from Nous Research. It keeps memory between sessions, can write reusable skills from tasks it has finished, and can talk to you in apps such as Telegram, Discord, Slack, WhatsApp, Signal and email. It is released under the MIT licence.
Hermes Agent builds up memory and skills over time, and its messaging gateway has to be online to reply. On a VPS the agent keeps running when your own computer is off, and its memory and skills stay in one place on a server you control.
No. You get a clean VPS with full root access and install Hermes Agent with the official one-line install script, then pick a model with "hermes model". Our support team can help with the server side. We do not run or manage the agent for you.
The Hermes docs do not publish a minimum server size. When the model runs at a provider, the agent itself is light, so a small plan is a fair start and you can move up later. A model that runs on the server itself needs much more memory and usually a GPU, and our VPS plans are CPU servers.
Hermes Agent can use Nous Portal, OpenRouter, OpenAI, Anthropic and custom endpoints. The docs say the model needs a context window of at least 64,000 tokens. You bring your own account or API key and pay the model provider directly. WaseerHost only charges for the server.
Run "hermes gateway setup" and follow the prompts for the platform you want. The gateway then runs on your server, so the agent can reply while you are away from your computer.
Yes. Hermes Agent can run its terminal commands in a Docker container if you set the terminal backend to docker. That keeps what the agent does on the server in a sandbox. Install Docker on your VPS first.
Yes. Open a ticket at any hour and a support engineer will help with the server side: SSH access, the firewall, your domain and keeping the process running. For questions about how Hermes Agent itself works, the Hermes docs are the best source.
Full root access ยท Your own server ยท 24/7 support