Saltar al contenido principal
Model logo

API de SeeDance 1.0 Pro

    SeeDance 1.0 Pro forma parte de la familia de modelos SeeDanceVer todos los modelos SeeDance

    Toca para subirPNG, JPG, WEBP, GIF · Máx. 30 MB
    Toca para subirPNG, JPG, WEBP, GIF · Máx. 30 MB
    Salida

    Tu video generado aparecerá aquí

    What is SeeDance 1.0 Pro?

    SeeDance 1.0 Pro is ByteDance's frame-guided video model in the SeeDance 1.0 family. On UnificAlly, this route starts from an image: provide a first frame, a last frame, or both. A text prompt is optional and can describe the intended subject motion, camera behavior, and transition. At least one frame is required, so this route is better suited to controlled image-to-video and endpoint interpolation than prompt-only generation.

    The available API controls are explicit: 4–12 second duration in one-second increments, 480p, 720p, or 1080p output, six selectable aspect ratios, an optional fixed-camera flag, and an integer seed. ByteDance's Seedance overview describes the wider model family, while the UnificAlly parameter reference defines the request accepted by this endpoint.

    Key features of SeeDance 1.0 Pro

    First- and last-frame control

    Provide either endpoint or both. Using two images lets you define the opening and closing composition instead of leaving both ends to the prompt.

    4–12 second duration

    Choose any whole-second value from 4 through 12. Short drafts reduce iteration time; longer runs leave more room for a transition to develop.

    Three output resolutions

    Generate at 480p, 720p, or 1080p. Draft at a lower resolution, then use 1080p for a selected delivery render when the extra detail is useful.

    Seed and camera controls

    Pass a seed for controlled reruns and camera_fixed for shots where subject motion should not be accompanied by a large camera move.

    What SeeDance 1.0 Pro is best for

    • Product transitions

      Anchor the opening and closing product views, then describe the motion that should connect them.

    • Image-to-video shots

      Animate an approved key visual while keeping its initial composition available as an explicit input.

    • Storyboard interpolation

      Use two storyboard frames to communicate the desired endpoints of a short shot.

    • Stable-camera clips

      Enable the fixed-camera option for scenes that should read as subject animation rather than a camera move.

    • Multi-format delivery

      Choose among landscape, square, vertical, and wider aspect ratios for different placements.

    SeeDance 1.0 Pro use cases

    Animate a product still into a second approved packshot without asking the model to invent the ending composition. Turn an illustration into a short vertical clip for a social placement, using the first frame as the visual anchor. Build a transition between two storyboard panels and use the prompt to specify pacing and subject motion. For repeatable evaluation, keep the prompt, source frames, duration, resolution, aspect ratio, and seed recorded together; changing one variable at a time makes output differences easier to attribute.

    SeeDance 1.0 Pro API examples

    Submit work to /v1/tasks. The example uses both endpoints; a request may instead supply only one of them. See the full parameter reference for current validation limits.

    curl -X POST https://api.unifically.com/v1/tasks \
      -H "Content-Type: application/json" \
      -H "Authorization: Bearer YOUR_API_KEY" \
      -d '{
        "model": "bytedance/seedance-1.0-pro",
        "input": {
          "prompt": "A slow, stable transition as the package rotates toward camera",
          "first_frame_url": "https://example.com/opening.png",
          "last_frame_url": "https://example.com/closing.png",
          "duration": 8,
          "resolution": "1080p",
          "aspect_ratio": "16:9",
          "camera_fixed": true,
          "seed": 42
        }
      }'
    

    The response includes a task_id. Poll GET /v1/tasks/<task_id> or include a callback URL in your integration to receive the completed video asynchronously.

    Preguntas frecuentes

    Preguntas frecuentes sobre SeeDance 1.0 Pro

    SeeDance 1.0 Pro is a ByteDance video-generation model exposed on UnificAlly for image-guided video. It accepts a first frame, a last frame, or both, plus an optional text prompt.

    This API route is designed for frame-guided generation, not prompt-only text-to-video. At least one of first_frame_url or last_frame_url is required.

    Yes. Supply both first_frame_url and last_frame_url to define the opening and closing frames. The model generates the motion between them.

    Duration accepts whole-second values from 4 through 12. Resolution can be 480p, 720p, or 1080p.

    The API accepts PNG, WebP, JPEG, or GIF input up to 30 MB. The short side must be between 300 and 6000 pixels, with an aspect ratio from 0.4 to 2.5.

    Yes. An integer seed is available for controlled reruns, and camera_fixed can reduce camera movement when the shot calls for a stable viewpoint.