Seedance 2.0 Prompt Gallery

A high-performing prompt is structured: define the subject, action, camera behavior, visual style, and pacing. Seedance 2.0 responds best when intent is explicit instead of purely descriptive.

Assign each reference a clear role: images for identity/style, video for motion and camera rhythm, and audio for timing. This multimodal setup helps Seedance 2.0 produce more controllable, cohesive outputs.

Anchor the character with stable references and keep identity traits consistent (face, hair, outfit cues). Then iterate one variable at a time (camera, background, or motion) to preserve continuity across angles.

Yes. You can write prompts shot-by-shot with explicit framing and transitions. Seedance 2.0 is designed for stronger temporal coherence, making connected narrative sequences more reliable than single-clip-only workflows.

Use concrete camera language such as tracking shot, dolly move, low-angle reveal, orbit, or push-in. Pair camera direction with action timing to improve motion readability and cinematic rhythm.

Start from a gallery pattern, generate a short draft, and refine in focused passes: first continuity, then camera behavior, then pacing and audio sync. This staged loop reduces random drift and speeds up final delivery.