Getting Started
Overview
What Multi-Router is, and how its parts fit together.
Multi-Router is a platform for sending requests to AI models from many providers through one API, and for managing the routing, prompts, credentials, spend and records of those requests. Applications send OpenAI- or Anthropic-format requests to Multi-Router with one API key, and routers decide which model and provider serve each request. Teams change models, prompts and providers on the router, with no change to application code, and see every request in Traffic.
How it works
A request reaches a model in four steps:
- An organization adds provider keys for the providers it uses, such as Anthropic, OpenAI or AWS Bedrock.
- It creates routers, each with the model, provider and settings for one job, such as a support bot.
- Applications call Multi-Router with an API key, naming the router in the model field.
- Multi-Router serves the request from the router's provider, and records it in Traffic with its cost.
What you can do
Multi-Router covers these areas, each with its own section:
| Area | Covers |
|---|---|
| Providers & Models | Provider keys, the model catalog, regions and variants, model groups and custom models. |
| Routing | Static, Adaptive, fallback and weighted routers, Claude, Codex and Gemini Proxies, aliases and router settings. |
| Prompts | Versioned prompts, written in Multi-Router or kept in GitHub, bound to routers. |
| Agents & Editors | Coding agents, editors and chat apps through Multi-Router. |
| API Access | API keys, scopes, router access and rate limits. |
| Observability | Traffic, request detail, request logging and cost. |
| Organizations & People | Accounts, organizations, people and roles. |
| Single Sign-On | Signing in through the company's identity provider. |
| User Provisioning (SCIM) | Managing people from the identity provider. |
| Self-Service Provisioning | Tools policies and the Portal, for setting up AI tools. |
| Machine Identity | Workloads that authenticate with their platform identity. |
| Budgets & Spend Control | Caps, notifications and blocks on spend. |
| Self-Hosting | Running Multi-Router in your own infrastructure. |
| API Reference | Every /v1 endpoint. |
Next steps
The Quickstart sends a first request in a few steps. Teams moving an existing application start with Using an SDK, and Key Concepts defines the terms used across these docs.