Documentation Index
Fetch the complete documentation index at: /llms.txt
Use this file to discover all available pages before exploring further.
Error format and the full list of codes.
status_code and request_id. Quote request_id when contacting support.
{
"error": {
"message": "model \"x\" is not available",
"type": "invalid_request_error",
"code": "model_not_found"
},
"status_code": 404,
"request_id": "req_…"
}
| HTTP | code | Meaning |
|---|---|---|
| 400 | model_required | The body is not JSON or has no model. |
| 400 | invalid_request_error | The model rejected the request; message says why. |
| 400 | context_length_exceeded | Prompt plus max_tokens exceed the context window. |
| 400 | streaming_required | The answer is too long — resend with "stream": true. |
| 400 | unsupported_parameter, unsupported_value | This model does not take that parameter or value. |
| 400 | content_policy_violation | The model refused the content. |
| 400 | wrong_endpoint | The model is served by another endpoint; message names it. |
| 402 | insufficient_balance | Not enough free balance. meta gives required_micro_usd, balance_micro_usd, held_micro_usd. |
| 404 | model_not_found | No such model in the catalog. |
| 404 | not_found | Unknown endpoint. |
| 404 | job_not_found | No such job for this account. |
| 409 | job_not_completed | The job has no result yet. |
| 410 | content_expired | The file is past its retention. |
| 413 | request_too_large | The request body is too large. |
| 429 | rate_limited | Too many requests — retry after Retry-After. |
| 429 | too_many_jobs | The account already has its limit of media jobs running. |
| 502 | backend_error | The model backend failed. Safe to retry. |
| 503 | no_provider_active | The service is temporarily unavailable. |
| 503 | model_unavailable | The model cannot serve this configuration right now. |
429 — after Retry-After; 502 — right away or after a short delay; 503 — later or with another model. Do not retry 400 and 404 without changing the request.