Enterprise · Mobile Device Management

Your organization's AI. Deployed your way.

Distribute GNTK/AI across your fleet with full control over API keys, model routing and usage policies. Mobile Device Management and Managed App Configuration out of the box.

Request a demo See the architecture
18AI models
3platforms
MDMMobile Device Management
Mobile Device Management Managed App Configuration Custom branding Identity provider integration Dedicated infrastructure White-label build (roadmap)
Two deployment modes

Centralized control. Zero-network option.

Remote mode routes through your managed AI Orchestrator — you control the keys, models and costs. Local mode runs Apple Intelligence on-device: nothing leaves the device, no network required.

Remote (Orchestrated)

Prompts route through your organization's AI Orchestrator, on dedicated infrastructure. Rotate API keys via Mobile Device Management without re-publishing. Full token and cost visibility per user.

Managed keys Model routing Cost tracking
Remote mode

Local (On-device)

Runs Apple Intelligence directly on the device. Compliant with strict data-residency requirements. Works fully airgapped — no outbound traffic.

Data residency Airgap-capable No network
Local mode
Twenty models, one endpoint

Route any model to any team, from one backend.

Lock specific departments to specific models, or allow users to switch. All routing is configured in your AI Orchestrator — no app re-publish required.

Haiku
Sonnet
Opus 5.5
Gemini
GPT-6 Luna
GPT-6.1 Sol
GPT 4o
GPT 4o mini
Groq
Grok
Grok Build
Mistral
Cohere
Workers AI
Perplexity
Muse Spark
Gemma 4
GPT-OSS
+ Routing modes
Auto
Exacto
Admin control

Everything your IT team needs. Nothing they don't.

Deployment, key management, cost visibility and policy enforcement — all handled outside the app binary so changes don't require a new release.

🔑

Key management

Inject API keys via Managed App Configuration. Rotate at any time without touching the App Store or re-enrolling devices.

⚙️

Mobile Device Management

Deploy via any Mobile Device Management provider. Push config, enforce personas, and set defaults per device group.

🎨

Custom branding

The administrator sets your organization's brand name and accent color from the app, and every member sees it — header, home screen and sign-in. Applied at runtime: same binary for every customer, no rebuild. A dedicated build with your own icon and app name is available under agreement.

📊

Cost visibility

Every Remote response logs input tokens, output tokens and USD cost. Aggregate by user, team or time period from your admin dashboard.

🛡️

Identity provider

Corporate sign-in with Microsoft Entra ID (multi-tenant), alongside Apple and Google. Employees sign in with the account they already use at work.

🌐

Data sovereignty

EU and US routing available across the full provider chain. Choose where each model request lands — and prove it.

⚖️

Weighted usage

One premium chat costs 1 unit. Model Fusion costs 1 per model that runs (2 to 5); image generation 3 (Lite) or 6 (Pro); Agent 5; Pipeline 3; Deep Research 50 (standard) or 150 (max). Economy models cost nothing — when the pool runs out, the team keeps working.

Built for deployment at scale

The operational features that matter in a fleet.

Every capability listed here is live today. Android, Windows, a web admin console and a fully white-labeled build are on the roadmap.

⌂

Local storage

Conversations stay on the device. Cloud sync is opt-in and administrator-controlled — never on by default.

$

Tokens & cost

Every Remote response shows input and output tokens plus the USD cost. Full auditability at the session level.

⌕

Full-text search

Users can find any conversation by title or content. Locale-aware, handles accents across all supported languages.

↗

Share & export

Export conversations as Markdown to Notes, Mail, or any iOS Share target. No data leaves the device unless the user initiates it.

⌘

Six languages

English, Spanish, Portuguese (BR), Italian, French and German.

♿︎

Accessibility

VoiceOver labels, Dynamic Type, and full keyboard support. Meets Apple's HIG requirements — no custom accessibility work required.

Questions

Frequently asked questions

What is GNTK/AI Enterprise?

The organizational deployment of GNTK/AI: your organization buys seats, your people sign in with the corporate account, and you keep control of API keys, model routing and usage policies from your own AI Orchestrator.

How is the app distributed to employees?

Through your own Mobile Device Management provider (Intune, Jamf, Workspace ONE, Kandji and others) or directly from the App Store. Configuration and API keys are pushed via Managed App Configuration, so rotating a key never requires re-publishing the app.

Does any data leave the device?

Only what you decide. Remote mode routes prompts through your organization's AI Orchestrator, with EU or US routing across the provider chain. Local mode runs Apple Intelligence on the device with no outbound traffic, and conversations stay on the device unless cloud sync is enabled by the administrator.

How do employees sign in?

With Microsoft Entra ID (multi-tenant), Apple or Google. The administrator shares one invitation code to add members to the organization.

How is usage measured and billed?

Seats are bought per organization (5-seat minimum, from U$D 15 per seat per month) and self-managed through Stripe. Members share one premium pool with per-function weighting; every Remote response logs tokens and USD cost, so usage can be aggregated by user, team or period.

Can it run on dedicated infrastructure?

Yes, under agreement: isolated backend, database, access control, domain and keys for your organization.

How to get started

We set your organization up. You keep control of it.

Same code and same backend as the personal app. From U$D 15/seat/month; we'll walk you through the plans.

For business

Enterprise deployment

Your organization buys seats; your people sign in with the corporate account and share one usage pool you control.

  • We create your organization with the seats you need (5-seat minimum) and invoice you directly
  • You distribute the app through your own MDM (Intune, Jamf, Workspace ONE, Kandji…), or your people install it from the App Store
  • Corporate sign-in with Microsoft Entra ID; the admin shares one invitation code to add members
  • Shared premium pool with per-function weighting, your brand name and accent color inside the app, seats and payment self-managed through Stripe
  • Dedicated, isolated infrastructure — backend, database, access control, domain and keys — available under agreement
Talk to us
For developers

Orchestrator API

The same backend that powers GNTK/AI, available as a REST layer deployable on dedicated infrastructure.

  • One POST /chat endpoint, twelve model providers
  • Automatic fallback, session memory, web search and embeddings
  • Deep Research agent with multi-source synthesis
  • EU and US data sovereignty across the provider chain
  • Source code, schema and deploy scripts. Yours to run.
Contact GNTK

Looking for the personal app? Download GNTK/AI →