Best Hermes Agent Setup: 6 Builds Ranked, Step By Step

The best Hermes Agent setup is Hermes Desktop with a Nous Portal model as the main brain, a local Ollama model as the fallback, a plain-text memory vault and one dashboard to drive it all from.

That is the stack I run every day, and I call it the three-layer setup because it has a brain, a cockpit and a memory.

If that sounds like too much for day one, the free setup is my second pick and the one-click beginner setup is my third.

This is the operator walkthrough, so every setup below comes with the real commands, and I checked each one against hermes --help on my own install before writing it down.

Top 3 Picks At A Glance

• 🥇 The three-layer setup is the best Hermes Agent setup overall, because it gives the agent a brain, a cockpit and a memory.
• 🥈 The free setup is the best choice if you want to spend nothing, because it runs on free Nous Portal models or free Ollama cloud models.
• 🥉 The one-click beginner setup is the best starting point if you have never opened a terminal, because Hermes Desktop installs like a normal app.

How I Ranked These Hermes Agent Setups

This is my own first-person ranking, and it comes from what I actually run and what I've covered in my videos and guides.

It isn't a lab test, so treat it as an experienced operator's opinion rather than a measured score.

I judged every setup on four things.

• The first is how fast you get from nothing to a working agent.
• The second is what it costs to keep running each month.
• The third is whether the agent remembers your business between sessions.
• The fourth is whether it keeps working when you close the laptop.

What Hermes Agent Is, In One Minute

Hermes Agent is a free, open-source AI agent from Nous Research that can use tools, run commands and remember things.

Hermes doesn't come with one fixed model, so a "setup" really means five choices.

Part of the setup	What it is	My default
The install	The Hermes program itself	Hermes Desktop, or the official install script
The front end	Where you talk to the agent	Hermes Desktop and hermes dashboard
The main model	The brain that does the thinking	A Nous Portal model
The memory	What the agent knows about you	A plain-text notes vault plus built-in memory
The plugins	Extra powers you add on top	Two to four catalog plugins, no more

Every setup in this ranking is just a different mix of those five parts.

🔥 Want my exact Best Hermes Agent Setup setup, step by step? Inside the AI Profit Boardroom you get step-by-step video trainings, my installable Agent OS and four coaching calls a week with 3,400+ members building real automations. → Join the AI Profit Boardroom

The 6 Best Hermes Agent Setups, Ranked

1. The Three-Layer Setup (Best Overall)

This is the setup from my 3-Layer Hermes OS guide, and it's the one I'd build if I started again tomorrow.

Best for: anyone who wants an agent they still use in three months.

Cost: the software is free, and the brain is free while you stay on a free Nous Portal model.

Here are the steps in the order I'd do them.

1. Download Hermes Desktop for macOS or Windows from the Hermes website, or run the install script if you prefer the terminal.
2. Run hermes setup --portal to log in to Nous Portal, pick a model and set Nous as your provider in one go.
3. Install Ollama and pull one small local model, then run hermes fallback add and choose it as your backup brain.
4. Make a notes folder with three notes, which are an about-me note, a decisions log and a daily log.
5. Tell Hermes to read that folder at the start of every session and to log what it did at the end.
6. Run hermes dashboard to open the web dashboard, which runs on your own machine at port 9119 by default.
7. Add plugins last, one at a time, with hermes plugins search and hermes plugins install.

The terminal install command is this.

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup

On Windows without WSL, the official docs give a PowerShell version, which is iex (irm https://hermes-agent.nousresearch.com/install.ps1).

The memory step is the one most people skip, and it's the reason their agent feels generic.

Hermes already has built-in memory files, and hermes memory status shows you what's active.

The notes vault sits on top of that, and I use Obsidian for mine because it's free and it's just text files in a folder.

2. The Free Setup (Best For $0)

Hermes Agent costs nothing, so the model is the only part of the stack that can send you a bill.

Best for: anyone testing Hermes, and anyone running easy daily jobs like summaries and research briefs.

Cost: it's $0, with rate limits at busy times.

You have two free routes.

1. Run hermes model, choose Nous Portal and pick a model that is marked free.
2. Run hermes model --refresh if a free model you expected is missing, because the list is cached.
3. Run hermes fallback add and choose a second free model, so a rate limit doesn't stop your work.

The second route uses Ollama.

1. Install Ollama from ollama.com and make sure it's running.
2. Run ollama launch hermes and pick a cloud model when it asks.

Ollama's cloud models are free with token limits, so heavy users will hit a ceiling.

Free lists rotate, so don't build your whole business around one free model.

When I recorded my Solar Mini 4 walkthrough, the Nous Portal free list showed seven models, and some appeared in both a free and a paid version.

Always pick the entry marked free.

3. The One-Click Beginner Setup (Best For Beginners)

If the terminal scares you, this is your setup.

Best for: first-time users and anyone setting Hermes up for a non-technical teammate.

Cost: the app is free, and you can stay on a free model.

1. Download the Hermes Desktop package from the Hermes website.
2. On a Mac, open the DMG and copy Hermes.app into Applications.
3. On Windows, open the download with Windows App Installer.
4. Sign in to Nous Portal from inside the app and pick a model.
5. Type "test" and wait for a reply.

The Mac package is for Apple Silicon only, so an older Intel Mac won't run it.

My one-click Hermes setup guide covers the Ollama version of this route in more detail.

If something looks wrong later, hermes doctor checks your configuration and tells you what's missing.

🔥 Want me to walk you through this setup step by step? Inside the AI Profit Boardroom, there's a full Hermes Agent course plus four live coaching calls a week where you can share your screen and get your setup fixed. You'll be building alongside 3,400+ members. → Get access here

4. The Profile-Per-Model Setup (Best For Power Users)

This is where Hermes gets fun.

Best for: people who already use Hermes daily and want different brains for different jobs.

Cost: it depends on the models you choose, and it can reuse a Claude subscription you already pay for.

A profile is a separate Hermes home with its own config, keys, memory, sessions and skills.

1. Run hermes profile create coder to make a profile called coder.
2. Run hermes -p coder setup to give that profile its own model.
3. Run hermes -p coder to chat with it, or just type coder chat, because every profile becomes its own command.
4. Run hermes profile list to see everything you've made.
5. Run hermes backup before any big change, so you can roll back.

I keep one profile per model, which means swapping brains is a single flag.

To run Claude on a subscription you already own, there's an official plugin from Nous Research.

hermes update
hermes plugins install claude-subscription-directsdk
hermes profile create claude-opus

Then you set two lines in that profile's config.yaml, which are the provider claude-subscription-directsdk-experimental and the default model opus.

That plugin needs Hermes 0.21.4 or newer and a Claude Pro or Max subscription, and it's marked experimental.

My Hermes Claude subscription guide shows the whole install from start to finish.

One warning comes straight from the official docs.

You should never point two running agents at the same profile, because both write memory and they'll confuse each other.

If you're choosing between brains, my best Hermes Agent model ranking goes deeper.

5. The Always-On Server Setup (Best For Business)

A laptop agent stops when the lid closes, so a business needs Hermes running somewhere that never sleeps.

Best for: agencies and teams that want to message the agent from Telegram, Slack or Discord at any hour.

Cost: you pay for the server or for Hermes Cloud, plus whatever your model costs.

On your own Linux server, the steps are these.

1. Connect over SSH, because the docs warn that browser consoles from some hosts corrupt pasted commands.
2. Run the same install script as above, then run hermes setup.
3. Run hermes gateway setup and connect a messaging platform.
4. Run hermes gateway install to add the gateway as a background service.
5. Run hermes gateway start, then check it with hermes gateway status.
6. Run hermes cron create to schedule your first recurring job.

The dashboard binds to 127.0.0.1 by default, so reach it through an SSH tunnel or a private network such as Tailscale rather than opening it to the internet.

There's also an official Docker image, and the quick start is docker run -it --rm -v ~/.hermes:/opt/data nousresearch/hermes-agent setup.

If you don't want to manage a server at all, Hermes Cloud on Nous Portal deploys a managed agent for you.

When I covered it, a Cloud Agent needed $10 in credits or an active Nous subscription to deploy, and it scaled to zero when idle.

The command I'd memorise for this setup is hermes pause, because it's the emergency stop for cron jobs and new gateway turns.

My best Hermes Agent VPS breakdown covers the server choice itself.

6. The Local Private Setup (Best For Privacy)

If your data can't leave your machine, run the brain locally.

Best for: private files, client data and anyone who wants a setup with no meter at all.

Cost: it's $0 forever, and the price is paid in hardware.

The model I use on small machines is LFM2.5 2.6B from LiquidAI, which runs in 8GB of RAM.

ollama pull hf.co/LiquidAI/LFM2.5-2.6B-GGUF:Q4_K_M
cat > /tmp/Modelfile.lfm26 <<'EOF'
FROM hf.co/LiquidAI/LFM2.5-2.6B-GGUF:Q4_K_M
RENDERER lfm2
PARSER lfm2-thinking
EOF
ollama create lfm2.5-2.6b -f /tmp/Modelfile.lfm26
hermes profile create lfm
hermes -p lfm setup

The rebuild step matters, because the direct download broke agent loops for me until I swapped in Ollama's native LFM renderer.

Use that long address too, because the short lfm2.5 tag gave me a different model.

Once the profile exists, hermes -p lfm status should show the local model on the model line.

On my Mac it ran at about 140 tokens per second, and it handled five-step tool chains cleanly.

It also failed at building full web apps, so I keep a bigger brain one profile away.

My run Hermes free forever guide has the full config and my honest scorecard.

🚀 Ready to put Best Hermes Agent Setup to work in your business? Inside the AI Profit Boardroom you get step-by-step video trainings, my installable Agent OS and four coaching calls a week with 3,400+ members building real automations. → Join the AI Profit Boardroom

The Best Hermes Agent Setups Compared At A Glance

Rank	Setup	Best for	Cost	Hardest part
1	Three-layer setup	Most people	Free software, free or paid brain	Keeping the memory vault up to date
2	Free setup	Testing and light daily jobs	$0	Rate limits on free models
3	One-click beginner setup	First-time users	Free app	Nothing, which is the point
4	Profile-per-model setup	Power users	Depends on your models	Managing many profiles
5	Always-on server setup	Businesses and teams	Server or Hermes Cloud, plus the model	Server security
6	Local private setup	Private data	$0, plus hardware	Model quality on hard builds

The Hermes Setup Commands Worth Memorising

These are the commands I use most, and every one is in hermes --help.

Command	What it does
hermes setup	It runs the interactive setup wizard.
hermes setup --portal	It logs in to Nous Portal and sets it as your provider.
hermes model	It picks your default provider and model.
hermes fallback add	It adds a backup model for when the main one fails.
hermes profile create <name>	It makes a new isolated profile.
hermes dashboard	It starts the web dashboard on your own machine.
hermes gateway setup	It connects messaging platforms such as Telegram.
hermes memory setup	It picks an external memory provider.
hermes plugins install <name>	It installs a plugin from the catalog.
hermes doctor	It checks your configuration and dependencies.
hermes backup	It zips your Hermes home so you can restore it.
hermes update	It updates Hermes to the latest version.

Setup Mistakes I See Operators Make

The first mistake is installing ten plugins on day one.

Install one, test it, then add the next, because that's the only way you'll know which one broke something.

The second mistake is skipping memory.

Without a vault, you'll re-explain your business in every session.

The third mistake is having no fallback model.

Free models get rate limited, and a second brain keeps your jobs moving.

The fourth mistake is exposing the dashboard to the internet.

Keep it on 127.0.0.1 and tunnel in.

The fifth mistake is never backing up.

Run hermes backup before you update or change profiles.

🚀 Want the finished version instead of building it yourself? The AI Profit Boardroom includes my installable Agent OS, with Hermes profiles, the memory vault and the dashboard already wired together. You also get four coaching calls a week and 3,400+ members to help when you're stuck. → Join the AI Profit Boardroom

Related Reading

• The best Hermes Agent UI compares the front ends you can put on top of any of these setups.
• The Hermes Agent plugins marketplace walkthrough shows how to install plugins safely.
• My Herald OS walkthrough covers a free desktop you can wrap around Hermes.

Also On Our Network

• 🌐 How a business or agency should set up Hermes Agent
• 🌐 All six Hermes setups scored in a full verdict
• 🌐 The Hermes setup quick-start for an Agent OS
• 🌐 Which model to run in each Hermes setup, with real bench scores

FAQ: Best Hermes Agent Setup

What is the best Hermes Agent setup?

My pick is the three-layer setup.

That means Hermes Desktop or the official install script, a Nous Portal model as the main brain, a local Ollama model as the fallback, a plain-text memory vault and a dashboard.

It's my own ranking based on what I run daily, not a measured benchmark.

How do I install Hermes Agent?

You can download the Hermes Desktop package for macOS or Windows from the Hermes website.

You can also run curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash on Linux, macOS or WSL2, then run hermes setup.

Can I set up Hermes Agent for free?

Yes, because Hermes Agent is free and open source.

The model is the only layer that can bill you, so pick a model marked free on Nous Portal, use a free Ollama cloud model or run a local model.

What are Hermes profiles?

A profile is a separate Hermes home with its own config, keys, memory, sessions and skills.

You create one with hermes profile create and run it with hermes -p followed by the name.

How do I keep Hermes Agent running all day?

Install Hermes on a small Linux server and run hermes gateway setup to connect a chat platform.

Then run hermes gateway install and hermes gateway start so it runs as a background service.

Which local model should I use with Hermes Agent?

On an 8GB machine I use LFM2.5 2.6B through Ollama, because it handles multi-step tool calls well.

It isn't good at building full web apps, so I keep a bigger brain one profile away.

📺 Video notes + links to the tools 👉

🎥 Learn how I make these videos 👉

🆓 Get a FREE AI Course + Community + 1,000 AI Agents 👉

About Julian

I'm Julian Goldie, an SEO entrepreneur, author and founder of the AI Profit Boardroom, which has 3,400+ members.

I help business owners scale with AI agents, automation and SEO.

• I built Goldie Agency, a 7-figure SEO and link building agency based in Bangkok.
• I've grown a YouTube channel to 400,000+ subscribers.
• I wrote "Link Building Mastery", which is available on Amazon.
• I run GoldieBench, where I score AI models on real one-shot builds.

→ Get my best AI training inside the AI Profit Boardroom

Start with the brain, add the memory, then add the dashboard, and you'll have the best Hermes Agent setup.