Envoy AI Gateway 1.0 is now generally available, with stable APIs, OpenAI-compatible routing across 16 providers, MCP gateway support, multimodal endpoints, quota-aware routing, and observability integrations for enterprise AI traffic.
Tetrate and the Envoy AI Gateway project announced Envoy AI Gateway 1.0 as a generally available open source release for managing enterprise AI traffic.
The Envoy AI Gateway release notes state that version 1.0 is the project’s General Availability release, with stable APIs and features intended for production use. The project is built around Envoy Gateway and is designed to help organizations route, observe, and control traffic to large language model providers.
According to the official Envoy AI Gateway v1.0 release notes, the release includes an OpenAI-compatible API across 16 LLM providers, an MCP gateway, multimodal and audio endpoints, enterprise observability features, and multi-tenant routing. The project’s GitHub release for v1.0.0 also describes the release as General Availability and highlights quota-aware routing, OpenTelemetry and OpenInference observability, and MCP gateway routing.
Tetrate, in a PR Newswire announcement, described Envoy AI Gateway 1.0 as a production-mature open source AI gateway built on CNCF Envoy Gateway. The announcement names Bloomberg, Nutanix, Tetrate, and the broader Envoy community as backers of the effort.
A central feature of the 1.0 release is provider abstraction through an OpenAI-compatible API. The Envoy AI Gateway release notes say the gateway supports routing to 16 LLM providers, allowing applications to use a consistent interface while operators manage provider-specific routing and policy behind the gateway.
The GitHub release notes also point to quota-aware routing. In practice, this is intended to help route AI requests with awareness of provider limits or configured quotas, rather than treating every upstream endpoint as interchangeable.
The release also includes support for the Model Context Protocol, or MCP. The official release notes and GitHub release both describe MCP gateway capabilities, positioning Envoy AI Gateway as a traffic management layer not only for model APIs but also for tool and context integrations that AI applications may depend on.
For workloads that go beyond text, the Envoy AI Gateway release notes say version 1.0 includes multimodal and audio endpoints. That expands the gateway’s scope to AI applications using richer input and output formats.
Observability is another major focus of the release. The GitHub release says Envoy AI Gateway 1.0 includes OpenTelemetry and OpenInference observability support. The official release notes describe enterprise observability features, while Tetrate’s announcement frames the project as a way for enterprises to operate AI traffic with controls similar to those used for other production services.
Multi-tenant routing is also listed in the official release notes. That matters for organizations running shared AI infrastructure across teams or business units, where operators may need to apply different routing, policy, quota, or monitoring rules for different tenants.
ChannelLife, reporting on Tetrate’s release, corroborated the 1.0 launch and noted that Bloomberg is using Envoy AI Gateway in production. ChannelLife also reported that Nutanix is integrating the project into its Agent Gateway and Enterprise AI offerings.
As AI applications move from experiments into production, organizations face familiar infrastructure problems in a new context: routing, rate limits, observability, governance, and vendor flexibility. Envoy AI Gateway 1.0 addresses those concerns by applying gateway patterns to AI model and tool traffic.
The significance of the 1.0 milestone is not that it introduces a single new AI model or application, but that the project is presenting its APIs and feature set as stable enough for production adoption. The official release notes, GitHub release, Tetrate announcement, and ChannelLife coverage all align on that point: Envoy AI Gateway has moved from earlier development stages to a generally available release focused on enterprise deployment.
Tetrate and the Envoy AI Gateway project announced Envoy AI Gateway 1.0 as a generally available open source release for managing enterprise AI traffic.
Envoy AI Gateway 1.0 is now generally available The Envoy AI Gateway release notes state that version 1.0 is the project’s General Availability release, with stable APIs and features intended for production use.
The project is built around Envoy Gateway and is designed to help organizations route, observe, and control traffic to large language model providers.
Continue reading
