Every plan combines isolated European hosting with the managed services Hermes needs around it: model access, secure proxying, usage controls, channels, voice, persistent data, monitoring, tested updates and human support. You use the agent; we operate the stack.
Plan differences mainly define agent capacity, usage allowances and support level. Backup frequency, premium model usage and other add-ons are shown separately.
Each plan includes the managed Hermes platform. Plan differences set your agent capacity, usage allowance and support level. 15-day free trial, no credit card. Cancel anytime.
Scale up or down anytime from your dashboard. Changes apply in seconds — no migration, no downtime.
The same three layers — infrastructure, managed platform, and operations — answer most questions about what you're paying for and how it differs from alternatives.
No. Each plan includes an isolated Hermes runtime plus the managed platform and operations described on the landing page. A VPS is only the infrastructure layer — it gives you compute, memory, disk and an IP address, but you still have to deploy Hermes, connect models and channels, handle secrets, configure TLS, add persistence, set limits, monitor health, plan updates and test recovery.
Hermes Haven gives your agent a server, the services it depends on, and an operations team responsible for keeping the supported stack working.
The monthly price is not based on RAM and CPU alone. It also covers:
You can build these pieces yourself on a VPS. Hermes Haven removes that setup and the ongoing operations work — including when certificates expire, providers change, limits are reached, disks fill, or releases introduce compatibility changes.
Yes. Hermes is open source. Self-hosting means you own deployment, integrations, security, monitoring, updates, backups and incident recovery. Hermes Haven operates the supported stack for you, and is designed for customers who value a ready, maintained system and a clear owner when something breaks.
Agent capacity (CPU, RAM, storage, agent count), access to the managed platform, plan-specific usage allowances (credits, requests), maintenance, and the stated support level. Add-ons such as extra RAM, hourly backups or Haven Credits are billed separately and shown transparently on this page.
Qwen 3.6 35B is included free on every plan, running on our European GPU cluster. Premium models (GLM-5.2, GPT-5.5, Claude, and others) are available through prepaid Haven Credits. You can also bring your own API keys for any OpenAI-compatible endpoint — your keys, your models, your rules.
When Haven Credits run out, the agent automatically falls back to the included Qwen model so conversations never break.
You see your usage and remaining balance in the dashboard. When you exhaust your Base Credits, the agent warns you and pauses until the next billing cycle. When Haven Credits are exhausted, premium model requests fall back to the included Qwen model automatically. You can top up Haven Credits or upgrade your plan at any time.
Managed backup, restore and recovery workflows are available on every plan. Hourly backups are available as an add-on (€10/month) for customers who need more frequent snapshots. Baseline backup frequency and retention depend on your plan. You can manage and restore backups from the dashboard.
Hermes Haven deploys and tests Hermes and platform updates. You control your agent's configuration, personality, memory and skills — we handle the runtime around them. Breaking customer configuration is avoided where possible, and communicated when not.
Every plan includes Hermes-aware human support — people who understand the Hermes runtime, proxy, models, channels, storage and platform controls together. Support channels and response targets vary by plan: email on Starter, email + chat on Plus, priority on Pro, and SLA 99.5% with dedicated support on Enterprise.
15-day free trial. No credit card. Qwen 3.6 35B included.