Hermes Agent v0.22 is the newest big release of the free, open-source AI agent from Nous Research, and this page walks you through updating to it and switching on every headline feature.
Nous Research calls it "The Threshold Release", and the official release notes date it 13 October 2026.
A small patch, v0.22.1, followed on 14 October 2026, so the version you install today is v0.22.1.
A fresh desktop install now opens on a guided setup chat that runs on a free trial model, so you can talk to Hermes before you own an API key.
In my video I grouped the release into three moves: walk in, plug in and hand off.
I'll use that same order here, and I'll show you the exact clicks and commands for each step.
This is a news walkthrough built from the official release notes, so I'll tell you what Nous Research says each feature does rather than pretend I've stress-tested all of it.
What Hermes Agent v0.22 Is
Hermes Agent is an AI agent that runs on your own computer or server, uses tools, remembers what it learns and works through your chat apps.
It's made by Nous Research, and it's free and open source under the MIT licence.
Version 0.22 is the biggest window Hermes has ever shipped.
The release notes count about 24,200 commits, about 6,080 merged pull requests, about 7,180 closed issues and about 2,520 contributors since v0.21.0.
Nous Research says that is roughly 2.5 times the size of the v0.21 cycle.
The codebase also got about a third smaller, falling from roughly 1.06 million lines to about 698,000 lines with no change in behaviour.
What's New In Hermes Agent v0.22 At A Glance
This table is the whole release on one screen, grouped the way I explained it in the video.
| Move | Feature | What it does for you |
|---|---|---|
| Walk in | First-launch setup chat | A fresh desktop install asks you a few questions in a chat and sets itself up on a free trial model. |
| Walk in | One-click local models | Hermes installs llama.cpp and picks a model that fits your hardware, with no account and no key. |
| Walk in | Stable release channels | Installs follow tested releases by default instead of every commit on main. |
| Walk in | Signed desktop packages | The desktop app is a signed package that carries its own Python and Node and updates itself. |
| Plug in | Connectors | Hermes can act inside hosted apps such as Linear and Notion after you sign in once in your browser. |
| Plug in | 600-plugin catalog | Every listed plugin is pinned to a reviewed commit and installs from the desktop app or the CLI. |
| Plug in | Password-manager sign-in | The browser agent logs in with 1Password, Bitwarden or Hermes's own vault without the model seeing the password. |
| Plug in | Claude 5.5 and GPT-6 model wave | New Claude, GPT-6 and Gemini 4 models are pickable where the providers offer them. |
| Hand off | Bot Screen | A bot on a Linux gateway gets its own desktop that you can watch and take over. |
| Hand off | Reasoning replay and outage waits | Prior reasoning carries across resumes, and Hermes waits out short provider errors instead of ending your turn. |
| Hand off | One backend and one gateway | One process per machine now serves every profile instead of one per profile. |
🔥 Want my exact Hermes Agent v0.22 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
Before You Update: The Upgrade Notes That Matter
I always read the upgrade notes before I touch the update button, and you should too.
These are the ones most likely to surprise you.
- Managed installs now run on Python 3.14, and the package manager installs it for you.
- The Windows desktop app is now MSIX-only and needs Windows 11 22H2 or newer.
- The macOS app is built for Apple Silicon only in the default builds.
- One messaging gateway per machine is now the default, and
hermes updatefolds per-profile gateways into it when nothing conflicts. - Chats no longer reset on an idle or daily timer, so you type
/newor/resetwhen you want a fresh start. - Ended sessions that have been inactive for 90 days are now deleted at startup, and you can set
sessions.auto_prune: falseto keep everything. - Memory providers such as Honcho, Supermemory, Hindsight and Mem0 moved into the plugin catalog, and existing users are migrated with their data intact.
If you keep old chats for reference, set the auto-prune option before your first launch on v0.22.
How To Install Or Update To Hermes Agent v0.22
There are three routes, and you only need the one that matches how you run Hermes today.
Route 1: The desktop app
The desktop app is now a sealed, signed package that updates itself.
Open it, accept the update when it offers one, and you're done.
On Windows it's a signed MSIX for x64 and ARM64, delivered through App Installer or the Microsoft Store.
On a Mac it's a signed, notarised build for Apple Silicon.
If you installed Hermes from the Microsoft Store, the Store handles the update for you.
Route 2: The command line
On a command-line install, you run this one command.
hermes update
For a fresh install on Linux, macOS or WSL2, the official one-liner from the Hermes docs is this.
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
On native Windows, you run this in PowerShell instead.
iex (irm https://hermes-agent.nousresearch.com/install.ps1)
Route 3: Docker
Docker users pull nousresearch/hermes-agent:stable.
The release notes say the stable and latest tags now move only when a release is published.
Choose your update channel
Source installs now follow tested SemVer releases by default instead of tracking every commit on main.
For the bleeding edge, run hermes update --set-channel main, or use the channel switch under Settings, then About, in the desktop app.
Moving back to an older release is labelled a rollback, and Hermes never does it without asking you.
Step 1: Run The First-Launch Setup Chat
This is the "walk in" part.
Open the desktop app on a computer where you've never used Hermes, and it starts a short setup chat instead of a settings screen.
One card at a time, it asks for your name, an accent colour and a layout.
Then it asks which apps you want to connect and which plugins you want to install.
After that it gives you a quick tour and offers up to four first tasks picked for you.
The chat runs on a free trial model called nous/welcome, once per operating-system user, so it needs no API key.
If you'd rather skip it, click Skip setup.
If you want it back later, go to Settings → Run onboarding again.
Existing installs and remote backends over SSH, URL or Cloud don't get the setup chat automatically.
The agent also gained a tool called set_theme, so you can type "make it violet, dark mode" in any desktop chat and it restyles the app.
The free welcome model is a one-time desktop trial, and the CLI, the TUI and the gateway don't get a free model.
When you're ready, hermes auth upgrade or /login in a chat upgrades the free identity to a full Nous account and keeps your connected apps.
Step 2: Connect Your Apps With Connectors
This is where "plug in" starts.
Connectors let Hermes act inside hosted apps such as Linear and Notion through the Nous tool gateway.
You ask for an app, sign in once in your browser, and the tools are usable in the same turn.
You don't need a Nous account to start, because Hermes creates a free connectors-only identity the first time you use an app.
The desktop app has a new Connectors page that lists hosted apps and your MCP servers together, and it replaces the old MCP tab.
You can connect several named accounts for one app, such as a work account and a home account.
You manage them with /connectors in a chat or hermes connectors in a terminal.
If you don't want connectors at all, tools.connectors.enabled: false turns them off, and nous.guest: false turns off guest identities.
🔥 Want my Hermes setup done with you, step by step? Inside the AI Profit Boardroom, I've got a full Hermes Agent course plus the Agent OS, with video walkthroughs on profiles, models and plugins. You'll be learning alongside 3,400+ members with four coaching calls every week. → Get access here
Step 3: Install A Local Model In One Click
This is the other half of "walk in", because it lets you run Hermes with nothing leaving your machine.
Hermes now installs and maintains llama.cpp with the right engine for your hardware, whether that's CPU, CUDA, Metal, Vulkan or ROCm.
It rates every catalog model against your GPU and RAM and picks the highest-quality build that fits.
The release notes say it only recommends models that fit fully in GPU memory.
Here's how you start it.
- On Windows, macOS or Linux, open Settings → Providers → Local Models in the desktop app.
- In a terminal, run
hermes modeland choose Local models. - Pick the recommended model, and let the download run, because downloads pause and resume.
- Start chatting, because the local server starts and stops with Hermes.
The catalog includes Qwen3.8 27B, Qwen3.8 Flash Next, Qwen3.6 35B-A3B and DeepSeek V4 Flash.
You can also use any GGUF model from Hugging Face.
Local models are great for privacy and cost, but a small local model won't match a frontier model on hard work, so test it on a real task before you rely on it.
Step 4: Add Plugins From The 600-Plugin Catalog
The plugin catalog launched during this cycle and grew to 600 entries.
Each plugin is pinned to a reviewed commit, and you can install it from Hermes Desktop or the CLI.
You can browse by category and by star ranking, with a page for every plugin and every author.
The plugin pages on the Hermes docs site shelve them under Memory, Desktop, Platforms, Web, Tools, Voice, Automation and Models.
The memory providers moved here too, so Honcho, Supermemory, Hindsight, Mem0, OpenViking and Holographic now live in the catalog with their makers.
ByteRover and RetainDB stay bundled with Hermes.
I covered the earlier version of this in my Hermes Agent plugins marketplace walkthrough, and the same rule applies now.
Install one plugin at a time, and check its known-issue warning before you add the next one.
Step 5: Turn On Password-Manager Sign-In
The browser agent can now log into websites with logins from 1Password, Bitwarden or Hermes's own encrypted vault.
The model never sees the password.
If nothing is saved, you get a masked prompt in the CLI, the TUI or the desktop app to save it.
One-time codes are filled without entering the conversation.
Card and address fills at checkout need your confirmation every time.
Step 6: Pick A Model From The New Wave
v0.22 brings what Nous Research calls the Claude 5.5 and GPT-6 model wave.
The new names are Claude Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5, plus GPT-6 Astra, Sol and Luna, GPT-6.1 Sol and Gemini 4 Argon.
They're pickable on Nous Portal, OpenRouter and native routes where the provider offers them.
You can star favourite models, see a subscription running low with a chip such as "17% left · resets 3:00 PM", and see when a rate limit lifts.
You can also pick reasoning effort as part of choosing a model, for example with /model X --reasoning high.
If you're unsure which brain to choose, my best Hermes Agent model guide ranks the options I actually use.
Step 7: Set Up Bot Screen For Hand-Offs
This is the "hand off" part of the release.
A bot running on a Linux gateway can now have its own desktop that Hermes Desktop streams live.
When it hits a login, a two-factor prompt or a CAPTCHA, you take the screen, finish that step yourself and hand control back.
The bot then carries on with your session.
With a Docker, SSH or Singularity backend, the bot's screen, computer use and the browser all run inside the sandbox rather than on your host.
The release notes say Bot Screen is off by default and needs some host packages, so plan a few minutes for setup.
If you're new to Hermes bots, start with my Hermes Agent bot mode guide first.
Step 8: Check Your Profiles On The Single Gateway
Hermes now runs one messaging gateway and one desktop backend per machine, and they serve all of your profiles.
Before this, every profile ran its own process.
Creating a profile now makes its bots live without a restart.
Each profile still keeps its own credentials, allowlists, cron jobs and sessions.
hermes update migrates existing per-profile setups automatically, and hermes gateway status explains any refusal in plain words.
Step 9: Watch And Steer Your Subagents
Running subagents now stay visible above the composer with each child's task, elapsed time and live transcript.
You open the view with F6 in the CLI and Ctrl+T in the TUI, and it's there in the desktop app too.
Each child has Steer and Stop buttons.
/stop now reaches background children and hands back their partial work.
The Smaller Hermes Agent v0.22 Changes Worth Knowing
All user-facing text now runs through one translation catalog with full coverage in 16 bundled languages.
About 19 rarely used tools now load only when the agent searches for them, which roughly halves the desktop tool schema sent on every call.
Ctrl+C now clears your draft instead of killing the turn, and the desktop app has a new Simple mode that works like a plain chat window.
A cron run that never reached the model now re-runs after 5, 15 and 30 minutes.
What Changed In The Hermes Agent v0.22.1 Patch
v0.22.1 landed on 14 October 2026, the day after the main release date.
Nous Research describes it as a patch release with a handful of fixes for the Windows Store, plugin updates and more.
It adds a curated model list at the top of the CLI model picker.
It also adds catalog plugins, including a Korean desktop language pack and a Brazilian Portuguese language pack.
hermes update or the desktop updater brings you straight to v0.22.1.
🔥 Want help picking the right Hermes setup for your business? The AI Profit Boardroom has Hermes Agent training, the Agent OS and four weekly coaching calls where you can ask me which features to switch on first. → Join the AI Profit Boardroom
Related Reading
- Hermes Agent plugins marketplace, explained
- The best Hermes Agent model for each job
- How an AI SEO optimisation agency uses agents like Hermes
Also On Our Network
- 🌐 How a business would actually use Hermes Agent v0.22
- 🌐 Every Hermes Agent v0.22 feature, scored with a verdict
- 🌐 The Hermes Agent v0.22 quick start for your Agent OS
- 🌐 The models in the v0.22 wave, measured on GoldieBench
FAQ: Hermes Agent v0.22
What is Hermes Agent v0.22?
Hermes Agent v0.22 is a major release of the free, open-source Hermes Agent from Nous Research.
Nous calls it The Threshold Release, and the release notes date it 13 October 2026.
How do I update to Hermes Agent v0.22?
The desktop app updates itself, and the Microsoft Store version updates through the Store.
On a command-line install you run hermes update, and Docker users pull nousresearch/hermes-agent:stable.
Do I need an API key to start Hermes Agent v0.22?
You don't need one on a fresh desktop install, because the setup chat runs on the free nous/welcome trial model.
The CLI, the TUI and the gateway don't get that free model, so you still connect a provider or a local model for real work.
What changed in Hermes Agent v0.22.1?
It's a patch from 14 October 2026 with Windows Store fixes, a curated model list at the top of the model picker and new catalog plugins.
Is Hermes Agent v0.22 free?
Yes, Hermes Agent is free and open source under the MIT licence.
You only pay if you choose a paid model or subscription.
If you run an SEO business and want a second pair of eyes on where agents fit, you can book a free SEO strategy session with my team.
📺 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've built a 7-figure SEO and link-building agency, Goldie Agency, with 70+ people in Bangkok.
- I've grown a YouTube channel to 400,000+ subscribers.
- I wrote the book "Link Building Mastery", which covers nine link building strategies.
- I run GoldieBench, where I score AI models on real one-shot builds.
→ Get my best AI training inside the AI Profit Boardroom
Update first, read the upgrade notes, then switch on the setup chat, connectors, local models and Bot Screen one at a time, and you'll get the most out of Hermes Agent v0.22.
