What is Asynchronous Generation
For time-consuming image generation requests, the platform supports asynchronous generation mode: after submitting the request, a task ID is returned immediately. The client polls for the final result — no need to hold a long connection.
Async mode suits scenarios with longer generation times and avoids request timeout interruptions. It also lets you submit multiple tasks in batch.
Workflow
Async generation has three steps: submit → poll → download.
1. Submit the task
Send a POST request to https://api.tvtapi.com/v1/images/async with your API Key (Bearer auth) and request parameters:
curl -X POST https://api.tvtapi.com/v1/images/async \
-H "Authorization: Bearer sk-xxxxxxxx" \
-H "Content-Type: application/json" \
-d '{"model":"model-name","prompt":"description text"}'
The response includes a task_id — your credential for querying and downloading.
2. Poll task status
Poll the task status using task_id:
curl https://api.tvtapi.com/v1/images/async/{task_id} \
-H "Authorization: Bearer sk-xxxxxxxx"
Task states include: queued, in_progress, succeeded, failed. Poll every few seconds until the state becomes succeeded or failed.
3. Download the image
After the task succeeds, fetch the image via:
curl -L https://api.tvtapi.com/v1/images/async/{task_id}/content \
-H "Authorization: Bearer sk-xxxxxxxx"
This endpoint returns a 302 redirect to the image's actual storage URL. Each image has a download limit; the task expires automatically after the timeout.
Use on the Web
If you don't want to write code, use the image workbench at /en-us/user/image-studio directly. The workbench has a built-in async toggle — once enabled, it polls automatically and displays results without manual API calls.
Notes
- Async functionality requires an admin to enable it, and the user account must have the corresponding permission
- Tasks have a validity period; expired tasks cannot be downloaded
- If a task stays in
in_progressfor too long, the system will time out and mark it asfailed