indentia.ai

Pricing

Pay for the agents that earn their keep.

Four parts: a base, a price per user, a price per agent, and what your agents actually spend on the language model. No abstract "seats": you pay for what works.

Hosting models

Four ways to run Indentia.

You decide how far Indentia sits from your infrastructure. Each hosting model receives a custom quote based on your scale, region and compliance requirements.

Custom quote

Multi-tenant SaaS

Shared infrastructure with logical isolation per tenant. A fast route to start using Indentia.

Request a quote →Sovereign cloud options →
Runs on
ScalewayMicrosoft AzureKubernetes
On request

Single-tenant SaaS

Dedicated namespace, managed by Indentia. No shared compute with other tenants, for isolation requirements that don't accept a "logical" answer.

Request a quote →Sovereign cloud options →
Runs on
HetznerMicrosoft AzureKubernetes
On request

Self-hosted

Indentia runs inside your own environment: data centre, private cloud or EU-cloud. You connect your own token providers and LLMs.

Request a quote →
Runs on
Red Hat OpenShift Container PlatformKubernetes
On request

Self-hosted air-gapped

Fully isolated, no internet. Including a mirrored registry, offline models and an offline evaluation harness.

Request a quote →
Runs on
Red Hat OpenShift Container PlatformKubernetes

LLM tokens

Your models, your token bill.

Indentia is a platform, not a model vendor. You choose which model runs under the hood, and tokens get billed where they're actually consumed.

Self-hosted or bring your own model

Run local models in your own environment, or bring your own account with OpenAI, Anthropic, Gemini or Azure AI Foundry. Token costs are billed directly by your model provider; we only charge for the platform.

SaaS: tokens included

For SaaS, the included LLM tokens and additional usage are agreed in your quote.

Add-on services

Support where it matters.

On-premises licence

Perpetual or annual, with annual maintenance. For air-gapped and strictly sovereign environments.

Implementation & custom

Pilot of 2-4 weeks on 2-3 sources. Then a phased rollout with our solution engineers.

Premium support

24/7 SLA, dedicated solution engineer, priority on bugs and features.

Training

Workshops for knowledge workers, agent builders and compliance officers.