Seedance 1.5 Pro is Live Now!Try Now!
Home/Explore/Seedance Video Models/bytedance/seedance-v1.5-pro/video-extend-fast
video-to-video

video-to-video

Seedance 1.5 Pro Video Extend Fast

bytedance/seedance-v1.5-pro/video-extend-fast

Seedance 1.5 Pro Fast Video Extend turns short shots into longer clips with natural motion continuation and strong temporal consistency. Supports 4–12 s extensions, 720p/1080p output with built-in upscaling, and seed-reproducible results for shot matching. Ideal for ads, trailers, and short-drama beats. Production-ready REST API with fast response, no cold starts, and affordable pricing.

Hint: You can drag and drop a file or click to upload

Whether to generate audio.
Whether to fix the camera position.

Idle

Your request will cost $0.2 per run.

For $10 you can run this model approximately 50 times.

One more thing::

ExamplesView all

README

ByteDance Seedance V1.5 Pro Fast — Video Extend

Seedance V1.5 Pro Fast (Video Extend) extends an existing video into a longer clip while maintaining motion continuity, subject consistency, and scene coherence. Use a guiding prompt to control what happens next, choose duration and resolution, and optionally generate audio for the extended segment.

Why it stands out

  • Smooth continuation that follows the original clip’s motion rhythm and visual style
  • Prompt-guided extension to steer actions, mood, and camera behavior
  • Resolution control with a simple cost multiplier (720p vs 1080p)
  • Optional audio generation with transparent pricing impact

Key Capabilities

  • Natural motion continuation Extends movement in a way that feels consistent with the input clip.

  • Scene coherence preservation Keeps lighting, composition, and identity cues stable across the extension.

  • Prompt-guided direction Describe what should happen next without rebuilding the scene from scratch.

  • Optional audio generation Enable or disable audio generation (pricing differs).

Parameters

ParameterDescription
video*Input video (public URL or uploaded file reference, depending on client).
prompt*Describe what should happen next and what must remain consistent.
durationOutput duration in seconds, up to 12s (allowed values depend on the endpoint configuration).
resolutionOutput resolution: 720p or 1080p.
generate_audioWhether to generate audio for the output video (true/false).
camera_fixedWhether to fix the camera position (true/false).
seedRandom seed (-1 for random; set a fixed integer for reproducible results).

Prompt Tips

Write prompts like a continuation director’s note:

  • Action next: what happens after the input ends
  • Motion intensity: subtle / smooth / dynamic / dramatic
  • Camera behavior: fixed / slow pan / gentle zoom (if you care, say it)
  • Consistency constraints: “keep the same character, outfit, lighting, framing”

Example prompts:

  • Continue the scene with the woman giving a wink to the camera, keep the same lighting and framing, subtle motion, natural facial expression.
  • Extend with a slow zoom out to reveal more of the environment while keeping the subject centered and the mood unchanged.

Pricing

Reference pricing (generate_audio = true)

ResolutionPrice per secondExample (5s)Example (10s)
720p$0.04$0.20$0.40
1080p$0.06$0.30$0.60

Reference pricing (generate_audio = false)

ResolutionPrice per secondExample (5s)Example (10s)
720p$0.02$0.10$0.20
1080p$0.03$0.15$0.30

Best Practices

  • Use clean input videos (clear subject, stable exposure, minimal compression artifacts).
  • Keep prompts specific but short, and state what must stay the same.
  • Start with 5s to validate motion and framing before extending longer.
  • Use 720p for fast preview iterations, then switch to 1080p for final.
  • Set a fixed seed when you want consistent reruns while refining the prompt.

Use Cases

  • Ad creative extension (turn short shots into longer deliverables)
  • Social content variations (longer cuts from a base clip)
  • Storytelling continuity (extend key moments with controlled motion)
  • Marketing edits (extend hero scenes with stable aesthetics)

Notes

  • The input video must be accessible to the runtime (public URL or valid uploaded reference).
  • If motion drifts or changes too much, add stronger constraints: “keep face unchanged”, “keep camera fixed”, “same outfit and lighting”.
  • Capability and allowed duration values can vary by deployment; rely on the endpoint’s UI/validator for the exact supported options.

Related Models