Orux AI
Documentation

Async tasks

GET /api/v1/tasks/{task_id} — check the status of a long-running job.

GET/api/v1/tasks/{task_id}Bearer sk-app-…

Orux AI uses a unified async task model for video, music, and queue-based image jobs (Midjourney imagine/upscale/variation, Suno extend, etc.). Submit returns task_id, GET reports current status, terminal states fire webhooks.

Status values#

FieldTypeDefaultDescription
queuedstringAccepted, waiting for a worker.
runningstringWorker is processing.
successstringDone. result is populated.
failedstringUpstream or routing failed. error is populated.
expiredstringJob exceeded its TTL (24h default).

Response#

FieldTypeDefaultDescription
task_idstringEcho of the submitted id.
statusstringOne of the values above (uppercase on the wire: QUEUED / RUNNING / SUCCESS / FAILED / EXPIRED).
aliasstringThe short model id you submitted with.
upstream_modelstringThe actual upstream model the gateway resolved your alias to.
channelstringInternal routing identifier for the backend that served the job. Informational only — safe to ignore.
submit_timestringISO-8601 timestamp the job was accepted.
finish_timestringISO-8601 timestamp the job finished. Absent while still running.
duration_secintWall-clock seconds spent on the upstream. Absent until finished.
result_urlsarrayOn success: URLs of the generated assets (video / audio / image / etc).
error_codestringPopulated on FAILED — stable machine code.
error_messagestringPopulated on FAILED — human-readable description.

Examples#

Poll a task

curl
curl https://orux.top/api/v1/tasks/task_01HZX... \
  -H "Authorization: Bearer $ORUX_API_KEY"