Source: Envoy AI Gateway v1.0 release notes (now Agent Router)

Release notes for the first GA of Envoy AI Gateway (2026-06-23): Kubernetes-native unified OpenAI-compatible API over 16 providers, provider fallback, token quotas, InferencePool and MCP routing; project later renamed Agent Router.

Contents
  • What: official release notes
  • Publisher: Envoy AI Gateway maintainers (project now named Agent Router, under the Agentic AI Foundation)
  • Date: 2026-06-23 (read 2026-11-14)
  • Original: theagentrouter.ai/release-notes/v1.0/ (formerly aigateway.envoyproxy.io)

Summary

v1.0 declares the v1beta1 control-plane CRDs stable. The gateway is an additive layer on Envoy Gateway (built on Envoy Gateway v1.8.1, Envoy Proxy v1.38.1, Gateway API v1.5.1).

Key claims

  • One OpenAI-compatible endpoint across 16 providers with request/response translation.
  • Automatic provider fallback; InferencePool endpoint selection for self-hosted models via the Gateway API Inference Extension.
  • Rate limiting on model tokens and per QuotaPolicy, with injected backend rate-limit filters to respect upstream provider limits.
  • BackendSecurityPolicy injects upstream credentials (API key, AWS, Azure, GCP), so clients never hold provider keys.
  • MCPRoute aggregates MCP servers with CEL-based per-tool authorization; reasoning tokens are reported separately in GenAI metrics.
  • The project was renamed Agent Router (“same code, same maintainers”).

Feeds

LLM gateways

Suggest changes →