Run Your Own Hermes Agent
The free, do-it-yourself path: install and run your own Hermes AI agent on your machine or server, plus the managed hosts and one-click templates that will run it for you.
Hermes is a capable, open-source AI agent — it can search the web, remember things, work with files and data, and run tools on your behalf. This guide is the do-it-yourself path: install Hermes on your own machine or server, point it at an AI model, and you've got your own agent — free, and entirely yours. 🦒
Rather not run a server? You do not have to. Several companies will host Hermes for you from a few dollars a month, and there are one-click templates if you would prefer your own cloud box. Jump to Where to host it. This guide covers the install either way, because a managed host still leaves you to point Hermes at a model and wire up your tools.
What you'll need
- A computer or server you can run commands on — Linux or macOS works best (a small cloud VM is perfect).
- About 15 minutes and basic comfort with a terminal.
- An AI model key. We recommend OpenRouter — one key gives you access to dozens of models, and you only pay for what you use. You can also use your own OpenAI, Anthropic, or Google key if you have one.
Step 1 — Install uv
Hermes uses uv to manage its Python environment. Install it with one line:
curl -fsSL https://astral.sh/uv/install.sh | shClose and reopen your terminal afterward so uv is on your path.
Step 2 — Get Hermes
Clone the project and install it:
git clone https://github.com/NousResearch/hermes-agent.git
cd hermes-agent
uv venv venv --python 3.11
uv pip install -e '.[all]'The [all] part matters — it pulls in everything (web search, tools, and memory support) so your agent has its full toolkit from the start.
Step 3 — Connect an AI model
We've had the best results with Qwen3 235B through OpenRouter — it's smart, fast, and inexpensive — so that's what we run for hosted agents. Grab an API key from your OpenRouter account, then point Hermes at it.
Set your key as an environment variable:
export OPENAI_API_KEY="sk-or-...your OpenRouter key..."And in your Hermes config.yaml, use a custom (OpenAI-compatible) provider pointed at OpenRouter:
provider: custom
base_url: https://openrouter.ai/api/v1
key_env: OPENAI_API_KEY
default_model: qwen3-235b-a22b-2507Prefer a different model or provider? Hermes speaks to any OpenAI-compatible endpoint, plus Anthropic and Google directly. Swap default_model for any model on OpenRouter, or change provider to anthropic / gemini with the matching key. See Hermes' own configuration docs for the full list of options.
Step 4 — Talk to your agent
From the hermes-agent folder, start a chat:
uv run hermesThat's it — you've got a running agent. Ask it to search the web, summarize a page, or help with a task, and it'll use its tools to get it done.
Make it smarter (optional)
- Give it a memory. gbrain is a free, open-source memory layer that plugs into Hermes so your agent remembers context across sessions. Install it with
bun install -g github:garrytan/gbrainand connect it as a tool. - Add more tools. Hermes supports the Model Context Protocol (MCP), so you can bolt on extra capabilities — file access, fetching, time, and more — by adding MCP servers to your config.
Keeping it running
Running Hermes on your laptop is great for trying it out. To keep an agent online around the clock, run it on a small always-on server (a cheap cloud VM works) and start it as a background service. Maintaining a server, securing it, and keeping the model topped up is the part most people would rather not deal with, which is what the hosted options below are for.
Where to host it
Hermes needs to be running somewhere for your agent to be any use. Your own laptop is fine for trying it out, but it has to stay awake and online, so most people move it onto a small server. You have two routes: a managed host that runs it for you, or your own cloud server with a one-click template.
Managed — nothing to run yourself
| Option | What it is | Roughly |
|---|---|---|
| Nous Hermes Cloud | First-party, from the team who build Hermes. Currently in preview. | $10 minimum in credits, or an active Nous subscription |
| Agent 37 | Always-on instance with a browser dashboard, terminal and file browser. Cheapest way in. | From $3.99/mo (bring your own model key) |
| Hostinger Managed Hermes Agent | Visual setup with Telegram already connected and AI credits included, so there is no API key step. | $5.99/mo intro, renews $11.99/mo, paid upfront |
| Cloudways Managed AI Agents | Hermes on managed infrastructure from an established host. | See their pricing |
Your own server — one-click template
These give you a machine you own outright. The install is a click in a marketplace instead of the terminal steps above, and you finish the model and channel setup yourself.
- DigitalOcean Marketplace — a Droplet that comes up with Hermes already installed. You pay only for the Droplet.
- Hostinger VPS — a Hermes template in the application catalogue, on a plain VPS you control.
Agntable and DeployHermes are two more managed options we found live, if none of the above fit.
Before you pick one. We do not run any of these and we get nothing from them, so treat this as a starting list rather than a recommendation. Prices were checked on 31 August 2026 and will move, so confirm on their site. Two things worth weighing: the cheapest tiers usually mean bringing your own model key, and your agent will be holding that key — plus a wallet key if you run Stacker Agent — so pick somewhere you would trust with both. It is also worth checking which Hermes version an image ships, as some marketplace templates run a few releases behind.
Was this page helpful?