AI Gateway
CreatedAzure-Samples
Experiments with Azure API Management's AI capabilities, integrating with Azure OpenAI, AI Foundry, and more.
About
This repository explores the AI Gateway pattern through experimental labs, leveraging Azure API Management to handle AI services APIs with security, reliability, performance, operational efficiency, and cost controls. Primarily focused on Azure OpenAI, the labs offer step-by-step instructions with Jupyter notebooks, Python scripts, Bicep files, and Azure API Management policies to accelerate the experimentation of advanced AI use cases and facilitate the deployment of Intelligent Apps into production.
Key Features
- Model Context Protocol (MCP) experiments
- Backend pool load balancing with Bicep and Terraform
- FinOps Framework integration for AI budget management
- Token rate limiting and metrics emitting
- Agentic experiments with OpenAI Agents SDK and AI Agent Service
- 520 GitHub stars
Use Cases
- Implement cost control and budget management for AI usage.
- Secure and manage access to Azure OpenAI and other AI services.
- Experiment with AI agents and their integration with various tools and APIs.