Seedance 2.0 Is Now Live on World Model Hub

April 9, 2026: Seedance 2.0 goes live

On April 9, 2026, Seedance 2.0 went live on World Model Hub, giving creators and teams a more reference-led path for AI video creation. Instead of relying on prompt-only generation, Seedance 2.0 is built for workflows where text, images, source clips, and audio cues all help shape the result.

This matters for users who already have storyboards, approved stills, prior footage, music cues, or a branded visual direction. Seedance 2.0 is strongest when the job is not just "generate a clip," but "keep the right subject, style, rhythm, and scene identity intact across multiple shots."

What you can now do with Seedance 2.0 on WMHub

• Combine multiple inputs in one workflow with a text prompt plus up to 9 reference images, 3 reference videos, and 3 audio tracks
• Generate 4s to 15s clips for short-form ads, product stories, storyboard tests, and branded social edits
• Choose review-friendly output settings with 480p or 720p output and 16:9, 4:3, 1:1, 3:4, 9:16, 21:9, and adaptive aspect ratios
• Work from a more continuity-friendly setup when the same character, product, prop, or visual style needs to stay recognizable

Why Seedance 2.0 stands out

• Multimodal referencing: Seedance 2.0 is built for text, image, video, and audio inputs instead of a prompt-only workflow
• Steadier continuity: It is better suited to recurring characters, branded products, and scene identity that need to hold through multiple beats
• Edit- and extension-friendly direction: It is useful when you want to continue a shot, revise a sequence, or push an existing idea further instead of restarting from zero
• Stronger storyboard fit: It makes more sense for teams working from approved frames, moodboards, packaging stills, performance references, or source footage

Best fits on WMHub

• Product launch videos, premium ads, and branded short-form campaigns
• Storyboard-led concept trailers and previz
• Music-led, beat-driven, or performance-heavy clips
• Creator content and social variants that still need a stable visual identity

What this release adds to WMHub

• A stronger option for multimodal, reference-led video creation
• Better coverage between prompt-led generation and heavier continuity workflows
• A more complete video lineup for creators choosing among cinematic prompting, native audio, short-form storytelling, and reference-heavy control