One installation, many customers
Each customer is an application tenant with separate data, people, quotas and catalogue. No cluster per customer.
One installation, many customers, your own prices: what you get and where to start.
Ventic is built for those who sell AI access to others: every customer is isolated, every cost is accounted for and you set the price.
Each customer is an application tenant with separate data, people, quotas and catalogue. No cluster per customer.
A provider profile with the branding, domains, contacts and legal documents you supply. Open WebUI keeps its own branding and licence.
You choose models and prices. Every call is held first and settled at the exact cost after, so margin never depends on estimates.
Delegate administration to your resellers and pay commissions on settled usage. Reseller features unlock after company verification.
Card or bank-transfer top-ups, group wallets, invoices and per-tenant usage. You do not write a billing system.
Cloud providers, customer keys (BYOK) or your own GPUs connected outbound with the remote agent, no public ports. The reversible PII filter helps with compliance.
Either you install the platform, or you start from the service Netter already runs. The difference is who keeps the infrastructure up.
Ventic runs on your Kubernetes: you are the provider, with your brand, your customers and your price list. Hardware, hosting and AI consumption are yours.
Ventic already running, hosted and operated by Netter: no cluster to source. Dedicated installations and agents require an agreement with the supplier.
The prices published by Licman and llm.netter.io. In EUR, VAT excluded, one-month commitment.
Licman: price list as of 28 October 2026. llm.netter.io: plans read on 10 November 2026; per-model prices update every minute with the exchange rate and time bands, so they are only on the service. There is no public reseller price list: discounts and commissions on the service are agreed.
Talk to us about resellingWe help you install Ventic on your infrastructure and start directly from the workload you want to put in production.