Questions
Frequently asked questions.
How ZenithAI is deployed, where it runs, what happens to your data, how it is priced, and how it supports your obligations. Something not covered here? Write to sales@data.in.
Where does ZenithAI run?
Inside your environment: on GPU servers in your own data centre, or in private cloud infrastructure your organisation controls. It is not a public SaaS service — there is no shared multi-customer cloud behind it.
Is ZenithAI a public cloud or SaaS AI service?
No. ZenithAI is a private deployment on infrastructure you control — on-premises GPU servers or a private cloud tenancy. It is not a shared multi-customer cloud service and not a public SaaS subscription.
Does our data leave our infrastructure?
The core platform — chat, models, document retrieval, computation, document generation and OCR — processes your data entirely on your servers. Optional connectors such as web search, email and outbound calling reach external systems by their nature; each is a module your administrators can enable, restrict or disable, and outbound activity is logged.
Does ZenithAI support India’s DPDP Act, 2023?
ZenithAI’s architecture supports data-residency and data-minimisation objectives, including obligations under India’s Digital Personal Data Protection Act, 2023, because personal data can be reasoned over without leaving your infrastructure. It is an architecture that supports your obligations — not a compliance certificate; your notices, consent flows and policies still do that work. More on DPDP & enterprise AI →
Which AI models does ZenithAI use?
ZenithAI runs leading open-weight model families, organised into four tiers — Instant, Fast, Smart and Genius — sized to your hardware. Models are examples, not commitments: your deployment can adopt newer or different supported models over time without changing the user experience.
Can we bring our own models?
Compatible open models can be brought to your deployment, subject to compatibility validation for your hardware and the platform’s serving stack.
Can ZenithAI operate in a restricted or offline network?
The core platform is designed to operate without internet access, and internet-dependent modules can be disabled. Fully isolated (air-gapped) deployments, including offline authentication, are scoped per engagement — tell us your isolation requirements and we will configure and validate against them.
How is ZenithAI priced?
As an enterprise deployment: a one-time platform licence with annual maintenance, plus implementation and support. There is zero token billing and zero per-user metering on your own infrastructure — usage is bounded by your hardware, not a bill. Your total cost of ownership (hardware, licence, implementation, operations) is quantified transparently during scoping. See the commercial model →
Is there any per-user or per-token billing?
No. On your own deployment there is no per-user subscription and no per-token metering — usage is bounded by the hardware you run, not by a bill. The commercial model is a one-time platform licence plus annual maintenance and implementation.
What does implementation involve?
A typical engagement moves from discovery (use cases, data sources, security requirements) through hardware sizing and installation, knowledge onboarding and department workspaces, to governed rollout with training and support from the Data Ingenious team. See the deployment path →
Who builds and supports ZenithAI?
ZenithAI is developed by Data Ingenious Global Limited, an Indian enterprise technology company with more than two decades of building and operating secure, mission-critical platforms — 50M+ users served and 800+ enterprise customers across its platforms. About the company →
Still have questions?
Talk to the people who build it.
Enterprise enquiries reach the ZenithAI team directly — engineers included.