Simple, sovereign pricing
Start free and scale as you grow. Every plan runs in eu-north-1 with content-free logs.
Showing monthly pricing.
Pro
€99/mo
- 500M tokens / month
- Unlimited endpoints
- 99.9% uptime SLA
- €0.0008 / 1M overage
- 3 regions
- Usage threshold alerts
Scale
Custom
- Unlimited tokens
- Unlimited endpoints
- 99.99% uptime SLA
- Negotiated overage
- Multi-region
- Usage threshold alerts
Prices in EUR, excl. VAT.
How usage works
You’re billed monthly for your plan, plus any overage past its token allowance. The dashboard shows requests, p95 latency, error rate, and token usage in real time, so there are never surprises — and the in-app bell warns you before you approach your allowance.
Frequently asked
How is usage billed?
Each plan includes a monthly token allowance. Usage beyond it is metered as transparent overage at the per-1M rate shown on your plan. You always see live usage on your dashboard.
Where does my data run?
All inference runs sovereignly in eu-north-1 (Nordics), on hardware we control. Prompts and completions are never logged or stored — only token counts, latency, and cost.
Can I change plans later?
Yes — upgrade or downgrade anytime. Your plan governs your token allowance, rate limits, regions, and SLA; changes take effect immediately.
Is it really OpenAI-compatible?
Yes. Point any OpenAI SDK or client at your endpoint's base URL with your Hardhaus key — the same chat/completions and embeddings APIs, sovereign backend.
What's included in the Free plan?
1M tokens per month, one live endpoint, and the full platform — keys, dashboard, audit log, and the model catalog. No card required.