Ross ROSS = Recommend OSS · open-source software intelligence for agents

envoyproxy/ai-gateway

Manages Unified Access to Generative AI Services built on Envoy Gateway observed · 2026-08-28

github.com/envoyproxy/ai-gateway · homepage · Go · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

88/100

  • Activity 99
  • Release rhythm 98
  • Longevity 48
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 28.0
  • age_days: 681
  • days_rel: 12
  • days_push: 7
  • n_releases_24m: 15

Full methodology

Adoption not part of the score

1961 stars · 344 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Envoy AI Gateway is an open-source, Kubernetes-native gateway built on Envoy Gateway that manages traffic from application clients to Generative AI services. It provides unified routing across 16+ LLM providers and self-hosted models, with authentication, rate limiting, failover, MCP gateway support, and cost/usage observability.

Use cases

  • route LLM traffic to multiple providers like OpenAI, Anthropic, and AWS Bedrock through one gateway
  • add authentication and rate limiting to OpenAI API calls from my apps
  • set up automatic failover between LLM providers
  • track token usage and cost per team or user for GenAI services
  • expose self-hosted model serving clusters with fine-grained access control
  • put per-user identity on MCP tool calls at the gateway layer
  • unify access to generative AI services in a Kubernetes cluster

When to choose

  • you run workloads on Kubernetes and want an Envoy-based, CNCF-aligned AI gateway
  • you need multi-provider LLM routing with failover, auth, and usage limiting in one control plane
  • you need enterprise observability and cost analytics for GenAI traffic
  • you serve both cloud LLM providers and self-hosted inference endpoints

When to avoid

  • you need a simple single-provider proxy without gateway features
  • you are not running Kubernetes or cannot operate Envoy Gateway
  • you want a lightweight client-side SDK rather than infrastructure

Facets

service · maturity stable

api-gateway routing auth rate-limiting middleware monitoring llm-inference large-language-models artificial-intelligence apis infrastructure-as-code cloud-computing go self-hosted cloud ai-gateway envoy envoy-gateway llm-routing genai mcp-gateway cncf observability failover cost-management devops kubernetes docker

10 sources

Member repositories

RepositoryRoleHealth v2
envoyproxy/ai-gatewaymain88

For agents

markdown · JSON · MCP: product_card(name="envoyproxy/ai-gateway")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem