How to configure your client

worouter is OpenAI-compatible. Any SDK, app or proxy that speaks the OpenAI API works by changing two values: the base URL and the API key. Pick your client below and copy a working snippet.

What you'll need

A workspace. Sign up when you don't yet have one.
An API key. Mint one on the API Keys page — it looks like wor_live_... and is shown once.
Your base URL:

https://<your-workspace>.ai.wozinga.com/v1

Append /chat/completions, /models, etc. to that base. Full API reference: /api.

Python Node.js Go Aider curl Cursor Continue.dev

Aider (AI pair programming)

Aider speaks the OpenAI API natively — export two variables and pass the model with an openai/ prefix:

# worouter is OpenAI-compatible: point aider's openai provider at it
export OPENAI_API_BASE=https://<your-workspace>.ai.wozinga.com/v1
export OPENAI_API_KEY=wor_live_...

# keep the openai/ prefix — the rest is the worouter model id
aider --model openai/anthropic/claude-haiku-4.5

The prefix tells aider to use its OpenAI-compatible client; everything after it is the worouter model id. Swap anthropic/claude-haiku-4.5 with any model from the catalog.

Anything else?

When your client lets you set a custom OpenAI-compatible base URL and an API key, it works with worouter. Tools known to work out of the box: LiteLLM, Open WebUI, LobeChat, BoltAI, llmcord, vLLM proxies, Helicone proxies, anything using openai or openai-node.

Stuck? Hit the WhatsApp button at the bottom-right and we'll help you wire it up.