04Deployment
Two paths in.
Both fully governed.
colliga.ai is designed to meet enterprise infrastructure requirements without compromise — whether your constraints are operational convenience or data sovereignty.
Standard
API-integrated enforcement
CaiGR™ governance runs in colliga.ai's managed cloud infrastructure. Your LLM calls are routed through the enforcement layer on every interaction — no installation, no infrastructure management, no operational overhead.
Fastest path to production — hours, not weeks
No model configuration changes required
SOC 2 Type II · data-in-transit encryption
Works with any cloud LLM provider
Best for: operational teams, rapid deployment, organizations without strict data-residency requirements
Sovereign
Self-hosted enforcement
The full CaiGR™ enforcement engine runs inside your own infrastructure. Your data never leaves your environment. Designed for regulated industries, air-gapped deployments, and organizations with strict data-residency or sovereignty requirements.
Full data sovereignty — no external calls
Deployable on-prem, in private cloud, or air-gapped
Works with local models (Llama, Mistral, etc.)
Audit log storage fully within your perimeter
Best for: financial services, healthcare, defense, regulated verticals, data-residency requirements
Shared across both deployments
GOVERNANCECaiGR™ deterministic envelope on every call
AUDITCryptographic audit trail, replayable on demand
COST53–58% lower token cost vs. post-hoc alternatives
MODELSModel-agnostic — works with any LLM