Universal AI Gateway
One governed API in front of cloud and local AI providers.
Interactive platform map
Architecture in context
The focused blueprint, its required foundation, and declared recommendations.
Universal AI Gateway
One governed API in front of cloud and local AI providers.
What this replaces
Applications become coupled to one model vendor, credentials spread across apps, and teams lose centralized visibility into spend and routing.
What your team gains
A consistent OpenAI-style API with centralized authorization, model abstraction, routing, cost controls and provider resilience.
What is inside the blueprint
How it fits the platform
Applications/agents -> LiteLLM gateway -> policy/routing/budget layer -> OpenAI, Anthropic, Google, Azure, Ollama/local endpoints and other providers.
LiteLLM, ai-stack-postgres
- AICORTEX Core PlatformThe gateway runs as a Core-managed service with Core's routing and persistence.
From prerequisites to operation
- Provider credentials or local model endpoints
- Database for gateway state
- Defined model naming/routing policy
- Deploy LiteLLM proxy
- Register providers and model aliases
- Create virtual keys/teams
- Enable spend logging and budgets
- Configure fallback chains and observability
- Model aliases
- Fallback/retry policy
- Budget/rate limits
- Caching and guardrails
- Logging callbacks
- User/project keys
- Provider availability
- Spend drift
- Rate-limit events
- Schema/version compatibility
- Gateway DB backups
What this unlocks with other layers
Private AI + Model Freedom
Users get one interface while builders can change providers underneath it without changing the experience.