Skip to main content
🎓 Claude Code Masterclass Learn AI-assisted development on Udemy — plus the companion book on Leanpub & Amazon. Start Learning
Nous Portal unifying model access, Tool Gateway, and Hermes Cloud hosted agent instances under one account
AI

What is Hermes Cloud? Running Hermes Agent Without a Server

Hermes Cloud runs a dedicated, always-on Hermes Agent instance for you — pick a model and server size, deploy in about a minute, billed hourly.

LB
Luca Berton
· 3 min read

Every Hermes Agent guide on this site so far — deploying it on Oracle Cloud’s free tier, wiring up Discord — assumes you’re running your own server. Hermes Cloud is Nous Research’s answer for when you’d rather not.

What Hermes Cloud Actually Is

Hermes Cloud runs a dedicated, always-on Hermes Agent instance for you, each with its own workspace and dashboard. Setup is reduced to picking a model and a server size in Nous Portal — no SSH, no installer, no systemd unit to write. Nous Research’s own announcement put the whole flow at “two clicks and 60 seconds” from empty account to a live agent.

Billing is hourly, drawn from your Nous Portal credit balance, and the instance scales to zero when idle — you’re not paying for a server sitting there doing nothing between conversations, which is a meaningfully different cost model from a VPS you’re renting whether you use it or not.

For teams, Hermes Cloud extends past a single instance: you can spin up separate agents per person or project, with granular access controls and billing consolidated under one org account rather than everyone managing their own.

Nous Portal Is the Account That Ties It Together

Hermes Cloud isn’t a separate product with its own login — it lives inside Nous Portal, the account layer that also covers:

  • Model access — a catalog of 300+ models across providers, reachable through one OAuth login instead of separate API keys per provider.
  • The Tool Gateway — web search (Firecrawl-backed), image generation, text-to-speech, and browser automation, each opt-in individually rather than all-or-nothing, billed through the same account.
  • Unified billing — one subscription and one invoice, instead of five different dashboards and top-up balances across providers.

If you’ve already run hermes setup --portal on a self-hosted instance (as covered in the Oracle Cloud deployment guide), you’re already using the same Nous Portal account — Hermes Cloud is just the option to run the agent itself on Nous Research’s infrastructure instead of your own.

New accounts also get a one-time credit grant usable specifically for Tool Gateway calls — not model inference — meant to let you try tool-using workflows (a web search, a page scrape) before committing spend to a paid plan.

Hermes Cloud vs. Self-Hosting: Which One to Pick

Hermes CloudSelf-hosted (e.g. Oracle Cloud free tier)
SetupPick a model + server size in Nous PortalProvision a VPS, run the installer, configure the gateway yourself
CostHourly, drawn from Nous Portal credit; scales to zero idleCan be $0/month on a free-tier instance, plus your time
Server managementNone — Nous Research runs itYou patch, monitor, and keep the gateway alive yourself
Where your data livesNous Research’s infrastructureA server you control
Team featuresPer-person/project agents, granular access controls, unified org billingYou build this yourself if you need it
Best forNot wanting to think about infrastructure at allFree-tier cost, full control over the box it runs on

Neither is strictly better — it’s the same tradeoff as managed vs. self-hosted anything. If the appeal of Hermes in the first place is running your own AI agent on your own infrastructure, self-hosting keeps that intact. If what you actually want is the agent without any of the server-admin part, Hermes Cloud removes that entirely, at the cost of paying for compute time and running on someone else’s servers.

Frequently Asked Questions

What is Hermes Cloud?

Hermes Cloud is Nous Research's hosted way to run Hermes Agent: instead of provisioning and maintaining your own server, you pick a model and a server size in Nous Portal, and a dedicated, always-on Hermes Agent instance with its own workspace and dashboard is deployed for you in about a minute. Billing is hourly, drawn from your Nous Portal credit, and the instance scales to zero when idle.

What is Nous Portal?

Nous Portal is the single account that ties together Nous Research's model catalog (300+ models from various labs), the Tool Gateway (web search, image generation, TTS, browser automation bundled behind one OAuth login), and Hermes Cloud. Instead of separate API keys and dashboards for each service, one login and one invoice covers all of it.

Is Hermes Cloud the same as self-hosting Hermes Agent on a VPS?

No — they're two different ways to run the same open-source agent. Self-hosting (for example on an Oracle Cloud free-tier instance) means you provision the server, run the installer, and manage the gateway yourself, but it can be free. Hermes Cloud is a managed instance Nous Research runs for you with hourly billing, no server management, but no free tier for the instance itself.

Does Hermes Cloud replace the free-tier VPS approach?

Not necessarily — it's a tradeoff, not a strict upgrade. A self-hosted instance on a free-tier VPS costs nothing but your own setup and maintenance time. Hermes Cloud costs money per hour the instance runs but removes the provisioning, patching, and process-supervision work entirely. Which one is worth it depends on how much you value not managing a server yourself.

What are the one-time free tool credits in Nous Portal?

New Nous Portal accounts get a one-time credit grant usable specifically for agent tool calls — things like web search or page scraping routed through the Tool Gateway — not for model inference itself. It's meant to let you try tool-using workflows without committing to a paid plan first.

#Hermes Agent #Hermes Cloud #Nous Portal #Nous Research #Hosting
Share:
AI Integration & GPU Platforms

Need help with AI Integration & GPU Platforms?

Need help deploying AI/ML platforms? Get expert consulting on OpenShift AI, GPU orchestration, and MLOps.

Learn more about AI Integration & GPU Platforms

Want to operate this yourself, in production?

Take the free AI Platform Engineer Readiness Scorecard to see which skills transfer — then build a production-shaped AI platform in the 4-week Bootcamp.

Take the Scorecard →
Luca Berton — AI & Cloud Advisor, Docker Captain

Luca Berton

AI & Cloud Advisor · Docker Captain · KubeCon Speaker

15+ years in enterprise infrastructure. Author of 8 technical books, creator of Ansible Pilot (1M+ YouTube views, 648K site users). Former Red Hat engineer. Speaker at KubeCon EU 2026 and Red Hat Summit 2026.

Free 30-min AI & Cloud consultation

Book Now