Question 1

What's the difference between Cloud and Private Deployment?

Accepted Answer

Same software. Cloud is hosted by us in the EU. Private Deployment runs in your data centre or your own cloud account: you operate it; we license the software, help you go live, ship updates, and can stay on for support. You bring your own model keys either way.

Question 2

Are there usage limits?

Accepted Answer

Cloud has a commercial request and retained-content allowance, content retention, and technical API limits, listed on this page. Technical limits return 429 when exceeded. Your configured budgets can also stop inference. Private Deployment has no commercial Cloud allowance; its configured technical limits and budgets still apply.

Question 3

What happens if we go over a limit?

Accepted Answer

If you exceed the commercial request or retained-content allowance for a month, we get in touch; for two consecutive months, we review your needs together. These commercial overages do not automatically stop inference. Technical rate and streaming limits return 429. When a configured budget runs out, requests through that budget stop until the next period or you raise the cap.

Question 4

Can I change later?

Accepted Answer

Yes. Talk to us if you want to move between Cloud and Private Deployment. If the Cloud price changes, it applies at the next billing period, with notice and a right to cancel.

Question 5

Does this product include usage of AI models?

Accepted Answer

No. Steinkauz AI is the policy layer. You connect your own model keys and pay your providers directly. That is how the product is meant to be used, on Cloud and Private Deployment. There is no included inference.

Question 6

How does pricing compare to using providers directly?

Accepted Answer

You pay Steinkauz AI for the policy layer. Inference is billed by your providers. There is no included inference budget.

Question 7

Can I use Steinkauz AI programmatically?

Accepted Answer

Yes. Cloud and Private Deployment expose supported OpenAI-compatible endpoints for chat, text embeddings and multipart transcription. Use API keys and concrete connection/model IDs. Upstream support, routing, optional budgets and request evidence apply to every request.

Question 8

How does Steinkauz AI control where my data goes?

Accepted Answer

Steinkauz AI does not decide the path. You do. With data routing policy, you allow or block sensitivity and provider-environment combinations. Your application declares request sensitivity; API keys carry a default sensitivity, a minimum sensitivity and a minimum provider environment. Incompatible requests are blocked before inference. See the Governance page for an interactive demo, or read the docs for the full setup.

Question 9

Is there a minimum number of organisation members?

Accepted Answer

No. Cloud is one organisation price, whether you have one member or many. Inviting people does not change what you pay.

Question 10

Can one person belong to more than one organisation?

Accepted Answer

Yes. Each organisation has its own subscription and Budgets. Entitlements follow the active organisation you select in the app.

Limit	Cloud
Governed requests / 30 days	50,000
Concurrent streams	5 per user (current API default)
API rate	30 requests per minute, per key (current API default)
Data retention	30 days by default for API content and snapshots; decision and usage ledgers have separate retention
Retained database content	5 GB of API content and evidence

Use Steinkauz Cloud, or run it on your own infrastructure.

Steinkauz Cloud

Private Deployment

What Cloud includes

Frequently asked questions

Start a Cloud trial, or talk to us about Private Deployment.