Context

Generation time depends on the model you choose, the length of your video, and current server load. If your generation seems stuck in “Processing” or “Pending” with no progress indicator, this article explains what to expect and when to take action.

Expected processing times by model

Higher resolution inputs and longer video durations increase processing time proportionally. A 2-minute video at 1080p will take significantly longer than a 15-second clip at 480p.

What affects processing speed

• Video length and resolution — Longer and higher-resolution videos require more compute time.
• Model choice — lipsync-2-pro uses diffusion-based super resolution, which adds processing time.
• Server load — During peak hours, jobs may queue before processing begins.
• Concurrency limits — Each plan has a maximum number of concurrent generations. If all slots are in use, new jobs wait in queue.

When to worry

• Under 5 minutes: Normal — wait for it to complete.
• 5–10 minutes: Check status using polling (GET /v2/generate/{id}) or webhooks.
• Over 10 minutes: The job may have hit an infrastructure issue. Do not resubmit the same request — this creates duplicate queue entries and slows processing further. Contact [email protected] with your generation ID.

How to check generation status

1. In the Studio — Your generation shows a status indicator (Pending, Processing, Completed, or Failed).
2. Via API polling — Send a GET request to /v2/generate/{id} and check the status field.
3. Via webhooks — Set up a webhook URL in your API request to receive a callback when the job completes or fails.

Tips to speed up processing

• Use lipsync-1.9.0-beta if speed matters more than quality.
• Trim your video to only the segments that need lip sync.
• Use 720p or 480p input instead of 4K when testing.
• Avoid resubmitting failed jobs repeatedly — each resubmission adds to the queue.

Related docs: Generation Times & Performance • Rate Limits • Troubleshooting