Kong Gateway

Kong Gateway puts API, LLM and MCP traffic behind a shared gateway on your own infrastructure. It's for platform teams that need consistent access controls and traffic policies across services and AI applications. The open-source gateway uses the Apache 2.0 license and runs natively on Kubernetes through Kong's official Ingress Controller.

Its core capabilities cover routing, load balancing, health checks and authentication. Teams can apply rate limits, handle TLS connections, and use plugins to modify requests and responses or add logging and monitoring. Plugins also let developers extend the gateway for their own services.

For AI applications, a common LLM API routes requests across providers including OpenAI, Anthropic, GCP Gemini, AWS Bedrock, Azure AI and Mistral. AI capabilities include semantic routing and caching, security checks and observability. Kong also governs MCP traffic, records analytics and can generate MCP interfaces from RESTful APIs.

The gateway can run on infrastructure you control, but requests to cloud model providers go to those services. Kong Konnect is a separate commercial cloud offering with a managed control plane, analytics, a service catalog and developer portals.

Deployment choices include running without a database and separating the control plane from the components that handle traffic. That separation lets teams manage gateway policies centrally while operating the traffic layer on their own infrastructure.

The Apache 2.0 license applies to the open-source gateway. Advanced AI and MCP plugins may require a commercial edition; check each plugin's deployment and license requirements.