Doubao-Seedance-2.0

Overview

The Doubao Large Model Team has launched a new generation of professional-grade multimodal creative video model, Seedance 2.0. It supports generating videos using images, videos, audio, and other multimodal inputs as references. Additionally, it features video editing and extension capabilities, can accurately reproduce various details while maintaining stable character characteristics, and offers extremely realistic audiovisual stability. It is deeply adapted to major core scenarios such as commercial advertising, film and television production, and social media marketing.

• Model type: Video generation
• Input: Video / Image / Text / Audio
• Output: Video
• Endpoints: video_generation

Capabilities

• 多模态参考:特征精细保持
• 视频编辑:精准定向修改
• 视频延长:无缝连贯叙事

Use cases

• Seedance 2.0 深度适配商业广告、影视制作与社交媒体营销三大核心场景。通过工业级标准的生成质量，抽卡成功率大幅提升，降低高品质内容的制作门槛与成本，优化从创意到成片的生产流程，为行业带来显著的效率提升。

Code example

Call with seedance-2.0-idle :

curl https://www.starunion.net/v1/video/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-idle",
    "prompt": "A silver wolf running across a snowfield under the moonlight, cinematic shot",
    "duration": 5,
    "resolution": "1080p",
    "ratio": "16:9"
  }'

# Video generation is async - poll with the id returned above
curl https://www.starunion.net/v1/video/generations/TASK_ID \
  -H "Authorization: Bearer YOUR_API_KEY"