Pricing
Priced per deployment.
CurionOS is installed into your cloud account and the agents are built with you, so the shape of a deployment differs every time. Tell us what you want automated and where it has to run, and we will put a number on it.
What every deployment includes
- Deployed into your own cloud account — single-tenant, your data never pooled
- Specialized agents built with you, each with its own private memory
- Shared organizational knowledge, with inline citations back to your documents
- Per-agent access control, so people reach only the agents they are added to
- Isolated Linux machines for work that needs a browser, a shell or files
- Approval holds on irreversible actions, with a full action log and kept artifacts
- Slack, Teams and agent email connectors, each authenticating the sender
- Your choice of model per agent, with no silent substitution
Questions
Why is there no price on this page?
Because the deployment is the product. What it costs depends on where it runs — your cloud account or ours — how many agents you need, which specialist skills we build with you, and how much model usage those agents consume. We would rather scope it with you once than publish a number that is wrong for almost everyone.
What is an agent, exactly?
A specialist with its own private memory that compounds with every task, access to the organizational knowledge you give it, and a real Linux machine to work on when a job needs a browser, a shell or files. You give it a name, a brief and the documents it needs.
Where does our data live?
In your own cloud account, in a single-tenant deployment: your database, your object storage, your memory namespaces. Each agent's memory is private to that agent and to the people you add to it; only organizational knowledge is shared between agents. Nothing is used to train a model.
Can we run it entirely inside our own AWS account?
Yes — that is the default shape, and the infrastructure is defined as code so the install is reproducible and reviewable by your own cloud team. The one thing to decide early is whether model inference may call an external API or must also stay inside your account; that changes the scope, so raise it in the first conversation.
Can we try it first?
Yes. We will set up a working pilot on your own documents so you can judge it on real work rather than a demo.
Still deciding whether this fits? info@curionos.com reaches us directly.