FAQ
Frequently asked questions
Onboarding, models, security, compliance, pricing, and exit, gathered in one place. Missing an answer? Write to us.
Getting started
Yes. The Shared tier has a free trial with 10M tokens included. You get an API key on the same day and can run against all 27+ open-source LLMs. Proof-of-concept migrates to Dedicated without code changes.
14 days from signature to production. API key within 24 hours after agreement is signed. Sign, Ready, Production, that is the whole onboarding flow.
Shared runs on pooled compute capacity with fair scheduling, good for PoC and development. Dedicated gives you guaranteed GPU / CPU allocation with 99.5% SLA and predictable latency, for production with business-critical workloads.
Models
27+ open-source LLMs out of the box, Llama, Mistral, Mixtral, Gemma, Phi, and more. All with public weights, no vendor lock-in. You can also deploy your own fine-tunes alongside.
Yes. We host weights in Hugging Face, GGUF, and safetensors formats and serve them via vLLM. Custom model deployment is included in Dedicated and above. We validate compatibility before go-live.
No. We have a zero-training policy. Your prompts, responses, and embeddings never become training data, neither for us nor third parties. This is documented in the DPA.
Security & data
On Danish soil. Danish AI supercomputer. Air-gapped from public internet. No connections to the US or China.
All of it. Prompts, responses, embeddings, vector indexes, audit logs, and model weights sit on Danish supercompute. Nothing leaves the perimeter.
Your team via API keys. Our SRE team has break-glass access, documented in the audit ledger on every use. No third parties. No US parent.
Compliance & contract
Our DPA is baseline GDPR with data-processor roles explicitly defined. The ISAE 3000 Type II report is provided to Private Cloud and Enterprise, it covers access controls, logging, change management, and sub-processors.
Pure OPEX. Monthly subscription to your partition, 12-month contract. No hardware investment. Token usage above the included volume is billed as overage, transparent and per model.
Shared runs best-effort. Dedicated: 99.5% uptime. Private Cloud: 99.9%. Enterprise: custom SLA negotiated per workload. Status page and incident RCA included for all tiers from Dedicated and up.
Exit-ready by design. You keep weights and data, export the audit ledger and fine-tunes, and close the partition. 30-day exit window included. No vendor lock-in, open-source models run anywhere else.
Yes, the Enterprise tier includes on-prem deployment. We deliver the same stack as an appliance or air-gapped deployment in your data centre. Defence and public-sector workloads with classified data typically run here.