AI-Gateway

AI-Gateway

Enterprise-grade AI model, tool, and agent management gateway based on Azure API Management.

MCPDevOpen source
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 1.0k
Source
mcp-github

Overview

AI Gateway is an enterprise-grade gateway powered by Azure API Management, designed for managing and controlling AI models, tools, and agents. It offers security, performance, observability, and cost control features, supports the MCP protocol, and is ideal for developers building production-ready AI applications. With this gateway, advanced functionalities such as model load balancing, semantic caching, and request routing can be achieved. Suitable for developers who need to deploy and manage AI models and tools in enterprise environments.

Capabilities

  • ▪OAuth 2.0 secure authentication
  • ▪Load balancing and semantic caching
  • ▪Token metrics and built-in logging
  • ▪Rate limiting and quota management
  • ▪MCP protocol support
  • ▪Multi-model routing

Use cases

Managing access to large language modelsImplementing model load balancingControlling token consumptionIntegrating real-time voice APIs

Setup

Requires: Python 3.12+uv 包管理器VS Code 和 Jupyter 扩展Azure 订阅
git clone https://github.com/Azure-Samples/AI-Gateway.git; cd AI-Gateway; uv sync; uv pip install -r pyproject.toml; code .

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How do I get started with AI Gateway?

Clone the repository, install dependencies, open VS Code, and select a lab.

Which protocols does AI Gateway support?

Supports the MCP protocol and HTTP protocol.

Related skills