Skip to content

SPECIALIST INTELLIGENCE. OPEN CHOICE.

The right model for the job. Your choice.

Some tasks need a specialist. Others need broader reasoning, a clear rule or a person. Cention brings them into one workflow, so you can choose what works for each step.

Illustrative workflow
  1. 01Understand
  2. 02Check
  3. 03Decide
  4. 04Execute
One case Context · Decisions · Ownership · History

CENTION’S OWN MODELS

Focused scope. Specialist capability.

We provide our own fine-tuned LLMs and ML models, designed to be especially capable within a defined scope. Their value comes from doing a specific job well.

You choose where to use them. Combine them with external models or your own, and change the mix as your needs evolve.

YOUR WORKFLOW

One process.
More than one kind of intelligence.

Cention fine-tuned LLM Specialist taskExternal or own model Your choiceML + business rules Defined decisionsPeople Judgement & approval
Illustrative combination · choose the participants for your process

MIX AND MATCH

Choose the intelligence. Keep the process together.

Model choice should help your business complete the work. The workflow coordinates the actions, and the case retains the context and decisions.

01

Use our specialists

Put Cention’s own fine-tuned LLMs and ML models to work on the focused tasks they are designed for.

02

Bring your preferred models

Use an external provider or your own model where it fits. Mix different models within a workflow.

03

Keep people and rules in the loop

Use explicit logic for predictable decisions. Bring in a person when the situation calls for judgement.

IN PRACTICE

A return request needs more than a good answer.

Interpret the request. Retrieve the order. Check the policy. Arrange the next action. Model outputs become useful when the workflow can act on them.

01 / UNDERSTAND

What is being requested?

An LLM interprets the request.

02 / CHECK

What does the policy allow?

Rules evaluate the conditions.

03 / RESOLVE

What needs to happen?

Connected systems act. People review exceptions.

Illustrative workflow. Model selection and actions depend on your process and connected systems.

HOSTED BY CENTION

Our models run with us. External models are your choice.

Cention is a cloud service. We host our own fine-tuned LLMs and ML models. Using them does not send model requests through an external model provider’s API.

You can opt in to an external provider — for example, by choosing OpenAI in a workflow. That choice enables the external processing required for that step. Connections to your own business systems remain part of the workflow you configure.

YOUR MODEL CHOICES

Cention-hosted by default.
External providers by opt-in.

Use our specialists. Bring your own. Choose another provider. Combine them where they fit.

Your data stays your data.

Your data never trains public models. SOC 2, GDPR, HIPAA compliant.

Explore the trust centre →

FROM COMPLEXITY TO PROGRESS

Put intelligence to work on your process.

Bring one process to your demo. We’ll map the manual steps, show where Cention can do the work and identify where your people add the most value.

Book a demo