> ## Documentation Index
> Fetch the complete documentation index at: https://docs.compute.prentis.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Errors

> Every error the inference API can return, whether it is safe to retry, and whether it costs you anything.

Every response carries an `x-request-id` header, including errors. Quote it when you
ask us about a call -- it is the only handle that reaches the whole path the request
took.

Error bodies follow the OpenAI shape:

```json theme={null}
{
  "error": {
    "message": "Invalid or revoked API key.",
    "type": "authentication_error",
    "code": "INVALID_API_KEY",
    "param": null
  }
}
```

Branch on `code`, not on `message`. Messages get reworded; codes do not -- they are
append-only, and a retired one is never reused.

## Does a failed call cost me anything?

Almost never. Of the 28 errors below, **2** still produce a usage record:

* `REQUEST_CANCELLED` -- Request cancelled by client.
* `PARTIAL_RESPONSE_TIMEOUT` -- The response was cut off because the model timed out.

The reason is the same in both cases: the model had already produced output when the
call came apart, so the tokens exist whether or not you received them. Everything else
on this page is free of charge when it fails.

## All errors

| Status | Code                           | Retry? | Billing           | Message                                                                                          |
| -----: | ------------------------------ | :----: | ----------------- | ------------------------------------------------------------------------------------------------ |
|    400 | `CONTENT_BLOCKED_INPUT`        |   no   | Not billed        | The request was blocked by the content safety policy.                                            |
|    400 | `CONTENT_BLOCKED_OUTPUT`       |   no   | Not billed to you | The response was blocked by the content safety policy.                                           |
|    400 | `CONTEXT_LENGTH_EXCEEDED`      |   no   | Not billed        | Request exceeds the model's maximum context length of {max_context_tokens} tokens.               |
|    400 | `INVALID_REQUEST`              |   no   | Not billed        | Invalid request: {detail}.                                                                       |
|    400 | `MEDIA_FETCH_FAILED`           |   yes  | Not billed        | Could not fetch media from the provided URL.                                                     |
|    400 | `MEDIA_UNSUPPORTED`            |   no   | Not billed        | Media type '{mime}' is not supported by model '{model}'.                                         |
|    400 | `MODEL_CAPABILITY_UNSUPPORTED` |   no   | Not billed        | Model '{model}' does not support '{capability}'.                                                 |
|    400 | `UPSTREAM_CONTENT_REJECTED`    |   no   | Not billed to you | The request was rejected by the model's content policy.                                          |
|    401 | `INVALID_API_KEY`              |   no   | Not billed        | Invalid or revoked API key.                                                                      |
|    402 | `INSUFFICIENT_BALANCE`         |   no   | Not billed        | Insufficient credits. Add credits to continue.                                                   |
|    402 | `PAYMENT_REQUIRED`             |   no   | Not billed        | A payment is overdue on this organization. Update the payment method to continue.                |
|    403 | `AGREEMENT_REQUIRED`           |   no   | Not billed        | Your organization must accept the current terms of service before using this API.                |
|    403 | `MODEL_NOT_ALLOWED`            |   no   | Not billed        | This API key is not allowed to use model '{model}'.                                              |
|    403 | `TENANT_SUSPENDED`             |   no   | Not billed        | Account is suspended. Contact support or check billing status.                                   |
|    404 | `MODEL_NOT_FOUND`              |   no   | Not billed        | Model '{model}' does not exist or is not available to this account.                              |
|    409 | `IDEMPOTENCY_CONFLICT`         |   no   | Not billed        | An earlier request with the same Idempotency-Key had a different body.                           |
|    413 | `MEDIA_TOO_LARGE`              |   no   | Not billed        | Media input exceeds the size limit ({limit}). Upload large files first and reference them by id. |
|    429 | `TENANT_BUDGET_EXCEEDED`       |   no   | Not billed        | Spending limit reached for this account.                                                         |
|    429 | `TENANT_RATE_LIMITED`          |   yes  | Not billed        | Rate limit exceeded for this account. Retry after {retry_after_seconds}s.                        |
|    429 | `UPSTREAM_RATE_LIMITED`        |   yes  | Not billed        | The model is temporarily over capacity. Retry after {retry_after_seconds}s.                      |
|    499 | `REQUEST_CANCELLED`            |   no   | **Billed**        | Request cancelled by client.                                                                     |
|    500 | `INTERNAL`                     |   yes  | Not billed to you | An internal error occurred. Reference: {request_id}.                                             |
|    503 | `DEPLOYMENT_SCALING_UP`        |   yes  | Not billed        | The deployment is starting up. Retry after {retry_after_seconds}s.                               |
|    503 | `STORAGE_UNAVAILABLE`          |   yes  | Not billed        | Service temporarily unavailable. Please retry.                                                   |
|    503 | `UPSTREAM_UNAVAILABLE`         |   yes  | Not billed to you | The model is temporarily unavailable. Please retry.                                              |
|    504 | `PARTIAL_RESPONSE_TIMEOUT`     |   no   | **Billed**        | The response was cut off because the model timed out.                                            |
|    504 | `REQUEST_DEADLINE_EXCEEDED`    |   yes  | Not billed to you | The request exceeded the maximum allowed duration.                                               |
|    504 | `UPSTREAM_TIMEOUT`             |   yes  | Not billed to you | The model did not respond in time. Please retry.                                                 |

<Note>
  `Retry? = yes` means the same request may succeed later with no change on your side.
  A `no` means something about the request or the account has to change first --
  retrying it unchanged will fail the same way.
</Note>