Skip to content

Support

How can we help?

Most questions are answered in the documentation or below. If yours is not, write to us.

Documentation

How to make a first request, choose a model, stream a response and handle errors.

Read the quickstart

Talk to sales

For volume pricing, invoicing, a security review or anything contractual.

Contact sales

Common questions

Getting started

How do I make my first request?

Create a key, then send a chat completion request to the API with that key as a bearer token. The quickstart in the documentation has copy-and-paste examples for curl, Python and TypeScript.

Can I use the OpenAI SDK?

Yes. Point the SDK at the Space Router base URL and use your Space Router key. Requests and responses follow the same format.

Where do I find a model's ID?

On the models page. Each row shows the ID to put in the model field of your request.

Billing

Why did my requests start failing?

The most common reason is that the account balance, or the budget on the key or team, has been used up. Add credits or raise the limit and requests resume.

How is the cost of a request worked out?

From the number of input and output tokens and the model's price per million tokens, as listed on the models page.

Keys and limits

What should I do if a key leaks?

Revoke it straight away and create a new one. A budget on every key limits how much a leaked key can spend before you notice.

Can I limit what one key may do?

Yes. A key can have its own budget, its own rate limit and its own list of models.

What does a 429 response mean?

The request was refused because a rate limit was reached, either yours or the model provider's. Wait briefly and try again, ideally with increasing delays between attempts.

When you write to us

  • Say which model you called and roughly when.
  • Include the error message or status code you received.
  • Never send an API key. If you think one has been exposed, revoke it first and then tell us.