ByteDance/

Seedance-2.0

Partner

$4.300

/ 1M tokens

*

A new-generation professional-grade multimodal video creation model developed, supports video generation with multimodal reference inputs including images, videos and audio.

Public

api versions

Input

Prompt

text prompt for video generation

You need to login to use this model

Login

Settings

First Frame ImageUpload Image File

Please upload an image file

Last Frame ImageUpload Image File

Please upload an image file

reference_images

You can add more items with the button on the right

reference_videos

You can add more items with the button on the right

reference_audios

You can add more items with the button on the right

Resolution

resolution of the output video

Aspect Ratio

aspect ratio of the output video

Duration

duration of the output video in seconds (4-15, or -1 for model to decide) (Default: empty)

Seed

random seed for reproducible output (Default: empty, -1 ≤ seed < 4294967296)

Watermark

whether to add a watermark to the output video

Generate Audio

whether the generated video includes audio synchronized with the visuals

Return Last Frame

whether to return the last frame image of the generated video

Output

Model Information

Dreamina Seedance 2.0 is a new-generation professional-grade multimodal video creation model developed by the Doubao Large Model Team. It supports generating videos using multimodal reference inputs such as images, videos, and audio, breaking the creative limitations of single-source materials. It also features video editing and video extension capabilities, enabling high-precision reproduction of object details, textures, timbres, visual effects styles, camera movements, and more. Character features can be stably maintained as well, giving creators director-level control over their work.

Dreamina Seedance 2.0 delivers ultra-realistic audio-visual stability: with exceptional motion stability and image detail, it produces footage comparable to real shooting and delivers a visually striking, indistinguishable-from-reality impact. It excels at handling complex scenes, vividly reproducing everything from subtle facial expressions and intense physical interactions to dynamic song and dance performances. It also features built-in professional camera movement, multi-shot narration, and text generation capabilities to enhance narrative tension. With native synchronized audio and video, it generates rich sound effects that precisely match visuals, and supports performances in multiple languages, accents, and dialects, endowing videos with ultimate completeness and immersion.

Dreamina Seedance 2.0 is deeply tailored for three core scenarios: commercial advertising, film and television production, and social media marketing. With industrial-grade generation quality, it greatly improves the success rate of content creation, lowers the barriers and costs of producing high-quality content, optimizes the production workflow from creativity to final video, and brings significant efficiency gains to the industry.

Multimodal reference-based with fine-grained feature preservation
- Video references
- Audio references
- Image references
Precise, targeted video editing
- Subject replacement
- Object-level editing
- Inpainting
Seamless video extension with temporal coherence
- Video extension
- Preceding scene generation
- Interpolation completion