Back to Models Guide
Bytedancevideo

Complete Guide to Using Seedance 2.0

Try This ModelTutorial

Overview

ByteDance Seedance 2.0 — next-gen text/image/video-conditioned video generation with synced audio, up to 15s, 1080p

Input Parameters

prompt
textarearequired

Describe the video (max 20,000 chars).

first_frame_url
file

Optional first-frame image (JPEG/PNG/WebP/BMP/TIFF/GIF, ≤30MB).

last_frame_url
file

Optional last-frame image (JPEG/PNG/WebP/BMP/TIFF/GIF, ≤30MB).

reference_image_urls
file

Up to 9 reference images (JPEG/PNG/WebP/BMP/TIFF/GIF, ≤30MB each).

reference_video_urls
file

Reference videos (MP4/MOV, ≤50MB each, combined ≤15s).

reference_audio_urls
file

Optional reference audio clips (WAV/MP3, up to 3, ≤15MB each, 2–15s each).

aspect_ratio
selectrequired

Frame dimensions. 'adaptive' lets the model pick based on inputs.

Options
16:99:161:14:33:421:9adaptive
Default: 16:9
resolution
selectrequired

Output resolution.

Options
480p720p1080p
Default: 720p
duration
sliderrequired

Video duration in seconds (4–15).

Min: 4Max: 15Default: 5
generate_audio
toggle

Generate AI-synced audio

Default: true
web_search
toggle

Allow the model to use online search during generation.

Default: false
NSFW Filter
toggle

Enable NSFW content filtering.

Default: false