Bytedance: Seedance 2.5

seedance-2.5

ByteDance's next-generation multimodal video generation, editing, and extension model with long-form, multi-asset, and multilingual support. Supports a Draft workflow: generate a 480p preview first, then render a 1080p final video from the completed Draft. Draft and final-render requests are billed ...

multimodalvideo editingvideo extension

Capabilities & limits

Model type: Video Generation
Input modalities
Output modalities
Video modes: Text to videoImage to video (first frame)Image to video (first & last)Multimodal reference

Updated Jul 31, 2026

Pricing

Pay as you go. Rates depend on the provider route you use; below is reference pricing when available.

Pricing by Video input and Video resolution
Video input	480p	720p	1080p
Without video input	$10.7 / per 1M tokens Output tokens	$10.7 / per 1M tokens Output tokens	$11.7 / per 1M tokens Output tokens
With video input	$6.4 / per 1M tokens Total tokens	$6.4 / per 1M tokens Total tokens	$7 / per 1M tokens Total tokens

All other configurations

$10.7 / per 1M tokens

Output tokens

One scenario is selected per request; prices from different rows are not added together. Cost equals measured usage multiplied by the selected rate.