AI Agent Hosting

Hermes Agent on your own server

Name: ZentisLabs Hermes Agent VPS
Brand: ZentisLabs

Run Hermes Agent, the open-source AI agent by Nous Research, on a cloud server with full root access. We install it at first boot with the official installer; you add your model provider and chat apps over SSH.

Deploy Hermes Agent See the setup

Hermes Agent by Nous Research · Open source, MIT

install: 1-click
from/mo: $84
GPU needed: No
locations: 19

Pricing

Choose your plan

Every plan comes with Hermes Agent installed at first boot, full root access and NVMe storage. Model usage is billed by your LLM provider.

Virtualized CPU: shared cores, the best value for one agent and its gateway.

AI Starter

One agent and its gateway

$84/mo

CPU: 4 vCPU
RAM: 8 GB RAM
Storage: 100 GB NVMe
Locations: 19 server locations
Access: Full root access
Hermes Agent: Installed at first boot

Deploy AI Starter

AI Business

Heavier tool use, several profiles

$158/mo

CPU: 8 vCPU
RAM: 16 GB RAM
Storage: 200 GB NVMe
Locations: 19 server locations
Access: Full root access
Hermes Agent: Installed at first boot

Deploy AI Business

AI Advanced

Teams, many gateways and subagents

$263/mo

CPU: 12 vCPU
RAM: 32 GB RAM
Storage: 300 GB NVMe
Locations: 19 server locations
Access: Full root access
Hermes Agent: Installed at first boot

Deploy AI Advanced

How It Works

Deployed by us, configured by you

No API keys or bot tokens at checkout. You enter them on your own server, where they stay.

Step 1
Deploy the server
Pick a plan and one of 19 locations. At first boot the official installer sets up Hermes Agent for a separate, non-root hermes user. Progress is logged on the server.
Step 2
Choose your model
Over SSH, hermes setup asks for your LLM provider and model. Your API key goes into the server's own config file; we never see it.
Step 3
Connect your chat apps
hermes gateway setup connects Telegram, Discord, Slack and more. hermes gateway install keeps the gateway running at boot and after you log out.

Setup over SSHbash

$ ssh root@YOUR_SERVER_IP# log in to your server
$ sudo -iu hermes# switch to the Hermes user
$ hermes setup# choose your LLM provider and model
$ hermes gateway setup# connect your chat apps
$ hermes gateway install# run the gateway as a service
$ hermes gateway start# start it, then message your bot

Commands from the Hermes Agent documentation. The login banner on your server shows them too.

Messaging Gateway

Talk to your agent where you already chat

The gateway is one background process that connects to every platform you configure. Platforms Hermes supports include:

Telegram
Discord
Slack
WhatsApp
Signal
Email
Microsoft Teams
Matrix
Mattermost
Home Assistant

The full, current list is in the messaging gateway docs.

Inside Hermes Agent

What the agent can do

One gateway, many chat apps: A single gateway process connects your agent to every messaging platform you configure, from Telegram and Discord to Slack and email.
Persistent memory: Hermes keeps curated memory across sessions in MEMORY.md and USER.md, and can search its past sessions.
Skills: On-demand knowledge documents the agent loads when it needs them. Hermes can create new skills from experience.
Scheduled tasks: Built-in cron scheduler: describe a task in plain language and Hermes runs it on schedule and delivers the result to a connected chat.
Subagents: Hermes can hand work to isolated child agents that run in parallel.
MCP and tools: Connect MCP servers over stdio or HTTP, and turn toolsets such as the browser on or off per platform.
Bring your own provider: Nous Portal, OpenRouter, OpenAI, Anthropic, Google Gemini, DeepSeek and many more, or any OpenAI-compatible endpoint.
Full root access: It is your server. Use the hermes CLI over SSH, install more software, and keep your data and keys on your own machine.

FAQ

Questions about Hermes Agent servers

What is Hermes Agent?

Hermes Agent is an open-source AI agent by Nous Research, released under the MIT licence. You use it in the terminal and, through its messaging gateway, in chat apps such as Telegram, Discord and Slack. It keeps persistent memory, uses and creates skills, runs scheduled tasks and can delegate work to subagents.

Is this an official Nous Research service?

No. ZentisLabs is an independent hosting provider and is not affiliated with Nous Research. We provide the server and install the open-source Hermes Agent on it with its official installer. Hermes Agent and its documentation are by Nous Research.

Which AI model does it use?

The one you choose. Hermes calls a hosted model through your own provider account: Nous Portal, OpenRouter, OpenAI, Anthropic, Google Gemini, DeepSeek and many more, or a custom OpenAI-compatible endpoint. The model needs at least 64,000 tokens of context. Model usage is billed by your provider, not by us.

Do I need a GPU?

No. The model runs at your provider, so Hermes only needs CPU, memory and disk for the agent itself, its tools and the gateway. Every AI plan is a CPU cloud server.

What do I set up after deploying?

Connect over SSH as root, run sudo -iu hermes, then hermes setup to choose your provider and model. To chat from your phone, run hermes gateway setup, hermes gateway install and hermes gateway start. Your dashboard, the server-ready email and the login banner on the server list the same steps.

Who can talk to my bot?

Only the people you allow. Without an allowlist the gateway denies every user who has not been paired by direct message. You set the allowed user IDs, for example TELEGRAM_ALLOWED_USERS, in ~/.hermes/.env on the server; hermes config set writes them there for you.

Does it keep running when I log out?

Yes. hermes gateway install sets the gateway up as a systemd service for the hermes user, and we enable lingering for that user at install, so the service starts at boot and keeps running after you log out. Opt-in Server Alerts email you if the server goes offline.

I use OpenClaw. Can I move to Hermes?

Hermes has a migration command: hermes claw migrate imports memory, skills, your persona and settings from ~/.openclaw. API keys are only imported when you pass --migrate-secrets. Run it with --dry-run first to preview. We also still offer OpenClaw servers.

How do I update Hermes?

Run hermes update as the hermes user. It updates to the latest release and reinstalls dependencies. hermes doctor checks the install and lists anything that is missing.

Run your own Hermes Agent

Your server, your model provider, your chat apps. Installed at first boot, configured in a few commands.

Get Started - $84.00/mo All servers