Agent Router
An open-source control plane for AI and agent traffic, built on Envoy, that gives one OpenAI-compatible API across hosted model providers, self-hosted inference and MCP servers while centralizing credentials, routing, quotas and failover.
Agent Router, formerly Envoy AI Gateway, is an open-source control plane for AI and agent traffic, powered by Envoy and Envoy Gateway. It gives application teams one consistent, OpenAI-compatible API for every model and tool, from hosted providers to self-hosted inference and MCP servers, while platform teams keep credentials, routing, quotas, failover and usage attribution in one place.
It can run as a standalone router on a laptop with one command, pointing any OpenAI-compatible client at a local endpoint, or be deployed on Kubernetes with Envoy Gateway using a two-tier pattern: a Tier One Gateway that handles authentication, top-level routing and global rate limiting, and a Tier Two Gateway that gives fine-grained control over self-hosted model access, including endpoint picking for inference optimization. It supports a wide range of AI providers, including OpenAI, Azure OpenAI, Google Gemini, Vertex AI and AWS Bedrock.
The CLI is called aigw, and the project keeps its CRDs, API group, container images, Helm charts and Go module path unchanged from its earlier name.