dreamina-seedance-2-5-260628
dreamina-seedance-2-5-260628 is the latest generation video generation model of the Dreamina Seedance series, which will be officially launched in August 2026. It supports Wensheng video and Tusheng video workflows, with rich input modes, including pictures, videos, audios, and pre-uploaded avatar materials. It supports first and last frame anchoring, flexible duration of 4-15 seconds, 480P and 720P output resolutions, multiple screen ratios, and optional audio synthesis. The model also supports network search enhanced generation, which can obtain the latest information through real-time network retrieval to assist video creation, and supports up to 9 reference pictures, 3 reference videos, and 3 reference audios. Seedance 2.5 further enhances multi-modal reference capabilities and generation quality based on 2.0.
- Context
- 0 tokens
- Max output
- 0 tokens
- Input
- image llm video
- Output
- video
Pricing
Input: $10.7 / 1M tokens
Output: $10.7 / 1M tokens
480p or 720p : $10.7 /
480p or 720p : $6.4 /
1080p : $11.7 /
1080p : $7 /
Endpoints
- openai