SeeDance 2.0 Pro API

ByteDance SeeDance 2.0 Pro - T2V, first & last frame, and omni-reference at 720p

·Features·FAQ

Documentation

Prompt

Text prompt. Use @Image1/@Video1/@Audio1 placeholders in omni_reference mode.

Upload images/videos above, then type @ in your prompt to reference them

Mode

Aspect Ratio

Only used for text-to-video mode

Duration

Video duration in seconds (4-15s)

Seed

Seed for reproducibility

Output

Your generated video will appear here

Features

What SeeDance 2.0 Pro API offers

Text-to-video, first and last frame, and omni-reference modes

Up to nine references in omni-reference (images, videos, and audio clips; audio capped around 15 seconds per docs)

4 to 15 second duration

720p output

Prompt placeholders such as @Image1, @Video1, and @Audio1 in omni-reference mode

Optional seed

REST API with JSON responses

Use cases

Built for

Primary

Brand spots: align multiple assets in one omni-reference prompt

Character continuity: slot references with named placeholders

Story beats: first and last frame for controlled transitions

Exploration: compare Pro against Fast with identical payloads

FAQ

About SeeDance 2.0 Pro API

Both expose the same modes, duration range, resolution, and nine-slot omni-reference in this playground. Pro is the higher quality tier; Fast targets shorter turnaround for similar inputs.

You upload up to nine files (images, videos, or short audio). The prompt references them with placeholders like @Image1 or @Video1 so each clause maps to a concrete asset.

Four through fifteen seconds using the duration slider.

They label the ordered reference list so the model ties each sentence in the prompt to a specific file slot.