One price per agent. A written exit.
A flat subscription, a 90-day pilot on your real data flows, and the freedom to leave at the end, with your data.
Talk to the teamFlat per agent. No tokens, no surprises.
We don't sell requests. We sell an agent that does the work, at a price your CFO can budget for.
From€750per agent per month
Flat subscription per agent, from €750 a month depending on the service level. One-off setup fee of €3,000 (dedicated infrastructure). No token billing, no surprises.
Start with a 90-day pilot| Level | Price |
|---|---|
| Pro | €750per agent per month |
| Enterprise | €980per agent per month |
| Setup | €3,000once, dedicated infrastructure |
Ninety days, on your real data flows.
Not a greenhouse demo: the agent works on your actual processes, and the result is measured in the audit chain.
From €5,250
90-day pilot, one agent
Setup of €3,000 plus three months at the Pro level (3 × €750).
- D0
Scoping.
Your use cases, your scope, and two or three indicators set down in writing.
- D1 to D90
The agent works.
On your real data flows. You follow its activity, its memory and its trust score in the cockpit.
- D90
Measure, then decide.
The indicators are read back from the audit chain. You continue, or you leave with your data.
You can leave on day 90.
The exit is written into the offer, not buried in an appendix.
No tacit renewal.
On day 90, the pilot only becomes a subscription if you decide so. Without your explicit agreement, it stops.
Your data comes back to you.
When you leave, your data and the memory the agent has built up are returned to you in a usable format.
Full reversibility.
No technical or contractual lock to take back control or change provider.
What's included.
- Scoping
- Use cases, scope and indicators, defined with your teams.
- Dedicated infrastructure
- One dedicated machine per agent, in Luxembourg. This is what the setup fee covers.
- The cockpit
- Live activity, memory, trust score and conversation with each agent.
- The sealed audit chain
- Every action timestamped and sealed, verifiable from the cockpit and exportable for your regulator.
- The procurement file
- Data processing agreement (DPA) and DORA contractual annex for your register of information.
The six governance documents are freely available, no form to fill in.
Read the System CardPublic funding may be available.
Depending on your company's size and the nature of the project, part of it may be eligible for Luxinnovation's Fit 4 AI or SME Package programmes. Eligibility is checked case by case, before any commitment.
Check my eligibilityThe questions your committee will ask.
- How is this different from ChatGPT or Claude?
- ChatGPT is reactive: it waits to be asked. An Easycore agent is autonomous: an AI you hand a job to, not a question. It sees when to act, acts in your tools and seals every action. On Monday at 9 am the fund report is ready: you read it.
- What about our regulator, or our auditors?
- Every action of the agent is timestamped, chained and sealed in the audit chain. You check its integrity from the cockpit and export the log for your regulator or your auditors. The six governance documents, including the System Card, are freely available for your file.
- What about our data?
- Each agent has its own dedicated machine, hosted in Luxembourg. Your data and your models cross no border, and you get them back when you leave, together with the agent memory.
- What if we want to leave?
- No tacit renewal: on day 90, without explicit agreement, the pilot stops. Your data and the memory built up are returned to you in a usable format.
- How much should we budget?
- A flat fee per agent per month, from €750 (Pro) or €980 (Enterprise) depending on the service level, plus a one-off setup fee of €3,000. No token billing, so no consumption drift. The pilot for one agent starts from €5,250.
- You are a small company. Will you be around in three years?
- Easylab AI is a Luxembourg company, based in Roeser, with products in production. And the exit is organised from the pilot onwards: whatever happens, your data and the agent memory remain yours.
- What if the agent makes a mistake?
- Your chain of responsibility does not change, it becomes traceable. Every action is traced and sealed, you follow the agent live in the cockpit, and its trust score tells you when to give it more.
Let's scope your pilot.
Thirty minutes to define the scope and the indicators. Reply within 24 business hours.