Skip to main content
2026/9/2 11:54 (UTC+8) · Docs Update · ByteDance 📊 When a Seedance call carries media, it is submission that is slow, not generation — ingest first and reference an asset:// asset ID Your media has to travel upstream to APIYI, then be forwarded to Volcengine and decoded and validated there; the task ID only comes back after all of that. With inline Base64 or a large image URL, the create-task call goes from about a second to tens of seconds — one customer raised their client read timeout to 300 seconds and still did not get a task ID. Generation itself normally takes 2–5 minutes, which is the provider’s normal speed and unrelated to submission. Referencing an asset:// asset ID drops the request body from megabytes to a few dozen bytes and returns the task ID immediately. Content checks on your media also move up to ingest time, so nothing fails halfway through a generation task, and asset IDs stay reusable — which also improves character consistency across shots. The asset library is free with the Seedance API, with no annual fee. A new page covers the latency breakdown, a comparison of the three ways to pass media, the three migration steps, and how to tell whether a task was created after a timeout: Asset-First: Faster, More Reliable Image and Video Inputs.
Back to Live Updates · 📚 Monthly archive