Video

Seedance 1.5 Pro

ByteDance's cinematic video model — smooth motion, first/last frame control, and reliable short-form clips.

About this model

Seedance 1.5 Pro is ByteDance's established short-form video model for text-to-video and image-to-video generation with smooth, cinematic motion. Its first/last-frame mode lets you anchor the opening and closing composition — ideal for ad storyboards where you already have hero stills from GPT Image 2 or Seedream on the same HiArt canvas. Clips run 4–12 seconds, long enough for social hooks and product demos without the complexity of multimodal reference pipelines.

Best for

  • Text-to-video from descriptive scene and camera prompts
  • First/last frame mode for controlled storyboard transitions
  • 4–12 second clips sized for Reels, TikTok, and product demos
  • Smooth motion with cinematic camera pans and zooms
  • Pairs with still frames generated on the same HiArt project
  • Lower complexity than Seedance 2.0 — fast to learn and deploy

Common workflows

  • Social short-form ads and organic video posts
  • Product demo loops from a single hero product photo
  • Storyboard animatics with defined start and end frames
  • Logo reveals and simple motion graphics concepts
  • B-roll generation to complement live-action edits

How to use on HiArt

  1. 01Add a video node on the HiArt canvas and select Seedance 1.5 Pro.
  2. 02Choose text-to-video, or upload first and last frame images from your still pipeline.
  3. 03Describe motion, camera movement, and pacing in the prompt.
  4. 04Set duration (4–12 s), generate, and iterate before moving to Seedance 2.0 for audio or omni-reference shots.

Capabilities

Provider
ByteDance Seed
Modes
Text-to-video, first/last frame
Duration
4–12 seconds
Audio
Not included — pair with Seedance 2.0 for synced sound

Learn more

FAQ

Does Seedance 1.5 Pro generate audio?
No. Seedance 1.5 Pro outputs silent video. Use Seedance 2.0 on HiArt when you need dialogue, ambient sound, or music generated in sync with the visuals.
How do first and last frames work?
Upload a starting image and an ending image. The model interpolates motion between them — useful when you already have key frames from an image model on the canvas.

Related models

Try on the canvas

Sign up free, open a project, and set model preferences before you generate.