An enterprise AI gateway for agents, models, and MCP servers
Tetrate Agent Router is one gateway between applications and every AI provider. Apps, teams, and agents connect to a single endpoint, where routing, credentials, budgets, guardrails, and telemetry are applied consistently, whichever model or provider serves each request. For personal use or enterprise-wide deployment, you decide. One gateway.
Connect to any provider and tool
One endpoint fronts every model provider and slots into the SDKs, agents, and observability stack already in use. These two pages are the fastest way to see what connects and how.
Supported providers
The 200+ models and provider families the gateway routes to, how each provider is authenticated, and where the live model catalog is published.
Browse providers →Integrations
Everything that plugs into the gateway: SDKs and frameworks, coding agents, guardrail vendors, observability backends, and MCP clients.
Explore integrations →Quickstart
Tetrate Agent Router is delivered in three deployment models, each with a quick start that ends with a working gateway and a first routed request. Ready to dive in?
Personal use. Multi-tenant, fully hosted by Tetrate, nothing to install.
Quick startFully Managed
Team-scale governance on a dedicated instance, hosted end to end by Tetrate.
Quick startSelf-Hosted Data Plane
Team-scale governance; AI traffic stays on the customer's own infrastructure.
Quick startUnsure which fits best? Which tier is right for you compares the tiers below, and the Where to begin page walks the decision with a flowchart.
Which Agent Router tier is right for you?
- Agent Router Service is suitable for personal use. Core AI model routing and management functionalities. Multi-tenant and fully hosted by Tetrate.
- Agent Router Enterprise, fully managed for multi-disciplinary teams and organizations that need advanced admininistration controls. Runs on a dedicated cloud instance, hosted by Tetrate.
- Agent Router Enterprise, self-hosted, where the layer that carries AI traffic runs on the customer's own infrastructure: prompts, responses, and credentials stay within the customer's domain.
In all forms, Tetrate hosts the configuration and management side.
How to choose
Two questions decide the tier:
- which capabilities are needed now?
- where should your AI traffic run?
The Service tier fits an individual developer who wants to start building right away. The Enterprise tier fits multi-disciplinary teams (developers, architects, operators, analysts, and AI leaders) who need to see and shape AI use across the organization: which models each team can reach, what each team spends, and which guardrails apply to every call. Both run on the same gateway, so moving up a tier is configuration, not migration: routing rules, API keys, and integrations carry over unchanged.
For a detailed comparison of both tiers, check the Tier Comparison page.
Documentation map
The documentation in six sections, mirroring the sidebar, with the most-used pages in each.