OpenAI-compatible AI gateway

One API.
200+ AI models.

Build with GPT, Claude, Gemini, and models from leading providers without maintaining a separate integration for each one.

OpenAIAnthropicGoogleMetaand more
quickstart.ts
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.ROUTERBASE_API_KEY,
  baseURL: "https://routerbase.com/v1",
});

const response = await client.chat.completions.create({
  model: process.env.ROUTERBASE_MODEL,
  messages: [{ role: "user", content: "Hello" }],
});

One integration

Build once. Choose the right model for every job.

01

OpenAI-compatible

Keep the SDK and request shape you already know. Change the API key and base URL to get started.

02

Unified model access

Use one catalog and one endpoint across supported providers, model families, and modalities.

03

Resilient routing

Build more dependable applications with smart routing and automatic fallback available behind the gateway.

Developer resources

Everything you need to make the first request—and the next hundred.

View all repositories

Ready when you are

Ship your first RouterBase request.

Pick a model from the live catalog, add your API key, and keep using the OpenAI-compatible tools you already know.