agentgateway vs HAProxy Ingress
agentgateway
agentgateway.dev
HAProxy Ingress
haproxy-ingress.github.ioSide by side
| Figure | agentgateway | HAProxy Ingress |
|---|---|---|
| Price from | Free | Free |
| Free plan | Yes | Yes |
| Free trial | — | — |
| Open source | Yes | — |
| Web | Yes | — |
| Windows | Yes | — |
| macOS | Yes | — |
| Linux | Yes | Yes |
| API | Yes | — |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 1 |
| Spec score · Kubernetes Ingress Controllers | 9.1 (#1) | 7.6 (#13) |
| Deployment | The project provides standalone binary, Docker and Kubernetes deployment paths. | The controller is installed and configured with a Helm chart, and the instructions require Helm version 3. |
| Purpose | Agentgateway is an open-source HTTP and gRPC gateway for routing, securing, observing, and governing service, LLM, MCP, and agent-to-agent traffic. | HAProxy Ingress configures HAProxy to route incoming requests from an external network to applications in a Kubernetes cluster. |
| Security | Listed security features include JWT and OIDC, API keys, authorization policies, TLS and mTLS with certificate rotation, and rate limiting. | The getting-started guide says HAProxy Ingress uses TLS SNI and the Host header to associate requests with ingress hosts. |
| Support | The enterprise page lists partner-provided commercial support, including a Solo.io offer with 24×7 coverage, upstream collaboration, and best-practice guidance. | The project lists a Kubernetes Slack channel, a users mailing list, and Stack Overflow for community discussion and questions. |
Plans
In their own words
agentgateway
Agentgateway is a free, open-source HTTP and gRPC gateway for routing, securing, observing, and governing service, LLM, MCP, and agent-to-agent traffic. Its LLM gateway offers an OpenAI-compatible API for providers including OpenAI, Anthropic, Bedrock, Gemini, Vertex, and OSS Llama runs. Self-hosted inference routing supports vLLM, TGI, and Triton, with choices guided by latency, cost, and model. For MCP it supports server discovery, scoped tool calls, and audit trails; A2A routing connects runtimes such as LangChain, CrewAI, and ADK. Security features include JWT and OIDC, API keys, authorization policies, TLS and mTLS with certificate rotation, and rate limiting. Guardrails inspect MCP calls and responses for prompt injection and unsafe payloads, while prompt guards support checks and redaction. Observability includes OpenTelemetry, Prometheus token-usage histograms, Jaeger traces, and request logs with latency, token counts, and realized USD cost. It can run as a standalone binary, in Docker, or on Kubernetes. Commercial support is listed through a partner.
HAProxy Ingress
HAProxy Ingress is a free, self-hosted controller that configures HAProxy to route external requests to applications in a Kubernetes cluster. It applies changes made in the cluster to HAProxy without requiring a separate update process. The project supports Ingress and Gateway API resources and describes more than 200 configuration options. Gateway API support is partial: the documented implementation covers GatewayClass, Gateway, TCPRoute, and HTTPRoute, but not HTTPRoute filters or resource status updates. Installation uses a Helm chart and requires Helm 3, a running Kubernetes cluster, and administrative privileges in that cluster. Version 0.16 requires Kubernetes 1.21 or newer. Local cluster options named by the project include minikube, kind, k3s, k3d, and colima. It associates requests with ingress hosts using TLS SNI and the Host header. Metrics can be collected with Prometheus and Grafana using Prometheus Operator. The project uses the Apache-2.0 license and lists community channels including Slack, a users mailing list, and Stack Overflow.