Glossary

Hybrid deployment

A hybrid AI deployment splits a system between your own environment and a cloud service. You choose which parts run where, and in the Ares hybrid shape the parts that handle your data run with you.

In a run

Your agents and data stay in your environment, and Ares sets their limits where they already run. Each call gets the same policy check, and goes on the same record, as it would in the cloud shape.

Our word for it

Your agents stay where they are.

What stays, and what moves

  • Stays: your agents, your data, your models and the record.
  • Moves: console sessions, and only if you allow them.
  • Never moves: prompts, model outputs and secrets.

Hybrid fits a team that already runs its agents on its own Kubernetes. The agents keep running there. Ares adds credentials, an access policy and the record around them, and nothing has to move. Your team keeps building agents the way it does now: see A2A for how they connect.

The same record as the cloud

The record in hybrid is the same as in the cloud shape, and it is kept in your environment. Export it and check it with the open verifier on a machine that is not ours. Residency routing applies in full, and the record shows where each call ran.

Where it stops

Hybrid needs a connection. Only the on-prem shape runs disconnected. People and agents sign in through your directory: see identity provider for which directories work today.

Compare the three shapes, or evaluate Ares with one of your agents in the hybrid shape.

Related terms

  • On-prem AIOn-premises AI is AI that runs on hardware inside your own building instead of in a vendor’s cloud.
  • Data residencyData residency is the rule that data must be stored and processed inside a set country or region.

Tell us about one workflow.

We will say whether an agent should do it, and what the first step costs.

A senior member of the team replies within two business days.