Envoy AI Gateway vs TensorZero
Envoy AI Gateway
aigateway.envoyproxy.io
TensorZero
tensorzero.comSide by side
| Figure | Envoy AI Gateway | TensorZero |
|---|---|---|
| Price from | Free | Free |
| Free plan | Yes | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | — | Yes |
| macOS | Yes | — |
| Linux | Yes | Yes |
| API | Yes | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 1 |
| Spec score · LLM Gateway Software | 8.4 (#8) | 8.0 (#13) |
| Spec score · LLM gateways | 8.4 (#10) | 8.6 (#7) |
| Spec score · Byok System | 8.4 (#8) | 7.9 (#15) |
| Deployment | The CLI can run locally on Linux and macOS, and production deployment uses Kubernetes with Envoy Gateway. | The gateway can be deployed with Docker, Docker Compose, Kubernetes using a reference Helm chart, or built from source with Cargo. |
| License | The project uses the Apache 2.0 license. | The FAQ describes TensorZero as open source under the Apache 2.0 License. |
| Observability | The CLI supports OpenTelemetry tracing and can record request and response details; the documentation describes configuration to hide input and output content. | TensorZero can store inference and feedback data for observability, optimization, evaluations, and experimentation in Postgres or ClickHouse. |
| Security | The project describes upstream authentication for securing connections to AI services and says provider credentials can be kept out of application workloads. | TensorZero's privacy policy says it uses physical, technical, organizational, and administrative security measures to protect personal data. |
| Support | The project directs users to its community Discord and GitHub issues for help. | The project README directs users to Slack or Discord and offers work teams a Slack or Teams channel by email. |
Plans
In their own words
Envoy AI Gateway
Envoy AI Gateway, now named Agent Router, is an open-source control plane for managing AI and agent traffic through Envoy. It provides an OpenAI-compatible API for hosted and self-hosted models and MCP servers, with integrations for providers including OpenAI, Azure OpenAI, Google Gemini, AWS Bedrock, Mistral, and Anthropic. Traffic controls include model virtualization, provider failover, token quotas, and usage-based rate limiting. Its MCP gateway supports server multiplexing, tool routing, OAuth authentication, and observability; InferencePool support adds routing and load balancing for inference endpoints. Credentials can be managed centrally, and supported short-lived credential flows include AWS Bedrock, Azure OpenAI, and GCP Vertex AI. The project describes metrics collection and monitoring for AI workloads. The free open-source project is listed at free. Its CLI supports local use on Linux and macOS; production Kubernetes deployment uses Envoy Gateway and requires compatible dependencies. The project uses the Apache 2.0 license and points users to Discord and GitHub issues for help.
TensorZero
TensorZero is a free, open-source stack that brings together an LLM gateway, observability, optimization, evaluation, and experimentation. Its gateway provides a unified API for model providers and supports routing, retries, fallbacks, prompt templates, schemas, multi-step workflows, and traffic routing for A/B tests. Documented integrations include Anthropic, AWS Bedrock, Azure, Google, OpenAI, OpenRouter, Together AI, vLLM, xAI, and OpenAI-compatible endpoints. Inference and feedback data go to the database you configure, where they can be queried with SQL; feedback can be tied to individual inferences or episodes. The gateway and UI can run through Docker, Docker Compose, Kubernetes and Helm, or a source build. API key authentication is available for gateway endpoints other than status and health checks. The project is available on GitHub, but is no longer maintained; its repository is archived and read-only. TensorZero is self-hosted and licensed under Apache 2.0. It collects pseudonymous aggregate usage analytics, excluding application data and inference inputs and outputs, with an option to disable collection.