--- title: "Using the STACKIT AI Model Service" description: "Learn how to deploy, configure, and monitor models with the STACKIT AI Model Service, from API integration to RAG patterns for sovereign AI systems." scfAsset: managed: false category: "guide" maintainers: - user: "can.celik1" external: true tags: ["STACKIT University", "Learning", "AI Model Service", "Inference", "RAG", "Deployment"] source_url: "https://framework.stackit.cloud/data-and-ai/assetcontainer/stackit/using-stackit-ai-model-service/" source_file: "docs/data-and-ai/assetcontainer/stackit/using-stackit-ai-model-service.mdx" --- ## Course Overview This course teaches the practical skills for deploying and integrating the STACKIT AI Model Service into your own applications. You'll start with the service's architecture and supported model formats, then move into deploying models to endpoints, configuring resources, and setting up versioning and rollback strategies with proper health checks — the operational details that separate a working demo from something you'd actually run in production. The second half is about keeping that deployment healthy and accountable once it's live: configuring authentication and access policies, issuing and scoping API keys, monitoring performance and usage metrics, and optimizing cost and resource utilization on STACKIT's managed, sovereign infrastructure. It's a focused, hands-on follow-up to LLM Fundamentals — less about the theory of the models themselves, more about the service operations you need once you're responsible for keeping them running. ### What You'll Learn - Navigate the STACKIT AI Model Service architecture and its supported model formats and tiers - Deploy models to endpoints with versioning and rollback strategies in place - Configure authentication, API keys, and access policies for production use - Monitor performance, usage, and cost, and troubleshoot with logs when something breaks ### Modules - Service Architecture and Capabilities - Model Deployment and Configuration - Access Management and Performance Monitoring