Seedance 2.0 ManualSeedance 2.0 Official Launch
🌀 Seedance 2.0 Video Is Live!
Ever since the days when you could only “tell a story” with text and a first / last frame, we’ve wanted to build a video model that truly understands what you’re trying to express. Today, it’s finally here.
Seedance 2.0 now supports four input modalities—image, video, audio, and text—so you can express more and control more.
You can use an image to lock in the visual style, a reference video to steer character motion and camera movement, and a few seconds of audio to set rhythm and atmosphere… combined with a prompt, creation becomes more natural, more efficient, and more like real directing.
In this upgrade, reference capability is the biggest highlight:
- đź“· Reference images can precisely reproduce composition and character details
- 🎥 Reference videos can replicate camera language, complex action rhythms, and creative VFX
- ⏱ Videos can be smoothly extended and connected—generate consecutive shots from prompts, not just “generate”, but “keep shooting”
- ✂️ Editing is upgraded too: swap characters, remove segments, or add content to an existing video
- We know video creation is never just “generation”—it’s control over expression. 2.0 isn’t only multimodal; it’s a truly controllable creative workflow.
- Seedance 2.0 multimodal creation starts here. Imagine boldly—leave the rest to it.