The Private, Secure
LLM for Your Business
Iris runs on private GPU infrastructure controlled by Vortex Dynamics — not OpenAI, not Google, not Anthropic. Your data never leaves our network, never trains a model, never gets shared.
AI that actually respects your privacy
Most AI tools send your data to third-party APIs. Iris doesn't. Every inference runs on hardware we own and operate.
End-to-End Private
Your prompts, files, and conversations are encrypted in transit and at rest. They never leave our infrastructure.
Private GPU Infrastructure
The model runs on dedicated hardware we control. No shared tenancy. No third-party API calls. No data leaving our network.
Zero Training on Your Data
We do not use your inputs or outputs as training data — ever. What you type stays between you and the model.
Built for Security Teams
Pentest questions, exploit research, red team workflows, CTF challenges — answered fully, without watered-down responses.
Request Beta Access
We're onboarding in small batches. Drop your email and we'll reach out with access details. Beta access is free while we finalise billing.
Already have an account? Sign in
Simple, transparent pricing
Choose the plan that fits your needs — upgrade or cancel anytime.
For individuals who need private AI
- 500 messages / month
- 32K context window
- Live web search & URL browsing
- File analysis — PDF, DOCX, XLSX, CSV
- Named AI memory (up to 5 contexts)
- 1 automated workflow
- Zero data sharing — ever
- Runs on private GPU
For power users & security professionals
- 1000 messages / month
- 128K context window
- Live web search & URL browsing
- File analysis — PDF, DOCX, XLSX, CSV, Markdown
- Named AI memory (unlimited contexts)
- 5 automated workflows with email delivery
- Full API access with personal key management
- Zero data sharing — ever
- Runs on private GPU
Custom deployments for organizations
- Unlimited messages
- 128K context window
- Live web search & URL browsing
- File analysis — all formats
- Named AI memory (unlimited contexts)
- Unlimited automated workflows
- Full API access with personal key management
- Dedicated support & SLAs
- Zero data sharing — ever
- On-prem or private cloud deployment
Why not just use ChatGPT?
| Feature | Iris | ChatGPT / Claude |
|---|---|---|
| Your data trains their model | Never | Yes (by default) |
| Prompts sent to third-party APIs | Never | Every request |
| Data sovereignty / Canadian hosting | ✓ | US servers |
| Custom AI memory (contexts) | ✓ | Limited |
| Automated workflows | 5 – unlimited | ✗ |
| API access with your own keys | ✓ | Paid tier only |
| Enterprise private deployment | ✓ | ✗ |
Product Roadmap
We're building Iris in the open with our beta users. Here's what's shipping next — and what's further out on the horizon.
Now
Private AI chat
Fully private conversations on our own GPU — no third-party API calls.
Live web search
Real-time search and URL browsing built into every conversation.
File analysis
Upload PDFs, DOCX, XLSX, CSV and ask questions about the content.
Named AI memory
Create persistent context windows that carry knowledge across sessions.
API key access
Bring your own integrations with personal API keys and usage tracking.
MFA & secure auth
Email-based multi-factor authentication and httpOnly session cookies.
Next
Upgraded LLM models
Continuous model upgrades as better open-weight models are released — you always get the best available.
Advanced document processing
Multi-document analysis, cross-file Q&A, automatic summarisation, and citation tracking.
Spreadsheet & data analysis
Ask natural language questions about tabular data, generate charts, and run calculations.
Scheduled tasks & automation
Set repeatable AI tasks on a schedule with email delivery of results.
Image understanding
Upload screenshots, diagrams, and photos for visual analysis and description.
Email & calendar integration
Summarise inboxes, draft replies, and manage calendar events through natural language.
Later
Team workspaces
Shared AI contexts, team memory, and collaborative task management for organisations.
Private on-premise deployment
Full air-gapped deployment inside your own infrastructure — your hardware, your network.
Plugin & tool ecosystem
Connect Iris to your internal tools, databases, and APIs via a plugin framework.
Voice interface
Talk to Iris hands-free — fully private speech-to-text and text-to-speech, on-device.
Real-time threat intelligence
Live CVE feeds, OSINT pipelines, and automated security briefings for security teams.
Multi-language support
Full interface and model support for French, Spanish, German, and more.
VS Code extension
Native Iris extension for VS Code — chat panel, inline code editing, tab autocomplete, and commit message generation, all powered by your private Iris instance.
Roadmap is subject to change based on beta feedback. Join the beta to influence what ships next.
Need a fully private deployment?
We can deploy Iris entirely within your own infrastructure — your cloud, your servers, your network. Air-gapped options available. Ideal for healthcare, legal, finance, defence, and government.