Errors and limits

Errors come back as JSON, with a status code and a message you can show.

JSON

{
  "error": {
    "code": "insufficient_balance",
    "message": "Your balance does not cover this request.",
    "type": "billing_error"
  }
}

Status codes

Status	Code	What to do
400	`invalid_request`	Fix the request. The message says what is wrong.
401	`unauthorized`	Check your key. It may be revoked.
402	`insufficient_balance`	Your balance does not cover the text. Add credit.
429	`rate_limited`	Wait for the seconds in `Retry-After`, then retry.
503	`unavailable`	The queue is full or the model server is down. Wait for `Retry-After`, then retry.

You are charged when the render finishes. A failed render is never charged, so retrying after an error costs nothing extra.

Limit	Value
Input	2,000 characters a request
Concurrency	2 requests at once per key, and 2 per account
Rate	60 requests a minute per key
Keys	10 active keys per account
Queue	Bounded across all accounts. When it is full, requests answer 503.