Saltar al contenido principal
Model logo

API de SeeDance 1.0 Pro Fast

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

    Modo
    Salida

    Tu video generado aparecerá aquí

    Actualizado

    What is the SeeDance 1.0 Pro Fast API?

    The SeeDance 1.0 Pro Fast API is live on UnificAlly as bytedance/seedance-1.0-pro-fast, from $0.007 per second of video. SeeDance 1.0 Pro Fast is the speed-oriented SeeDance 1.0 route for prompt-led generation. It exposes two modes on UnificAlly: text-to-video and image-to-video using a first frame. Unlike SeeDance 1.0 Pro, it does not accept a last frame, so choose it when quick prompt iteration or animating a single source image matters more than controlling the exact closing composition.

    The route supports 4–12 second videos in whole-second increments and 480p, 720p, or 1080p output. Text-to-video offers six aspect ratios: 1:1, 3:4, 4:3, 9:16, 16:9, and 21:9. Image-to-video uses the uploaded first frame as its visual source. Optional camera and seed controls let you specify a stable viewpoint and keep an evaluation setup reproducible. ByteDance publishes the wider family's background in its Seedance technical report announcement; the UnificAlly endpoint reference lists the live request fields.

    SeeDance 1.0 Pro Fast API specs

    Model IDbytedance/seedance-1.0-pro-fast
    EndpointPOST https://api.unifically.com/v1/tasks
    Request typesText-to-video, image-to-video from a first frame
    Duration4 to 12 seconds, any whole second
    Resolution480p, 720p, 1080p
    Aspect ratio1:1, 3:4, 4:3, 9:16, 16:9, 21:9
    Reference inputsOne first frame image (png, webp, jpeg, gif, up to 30 MB). No last frame
    Rate at 480p$0.007/sec
    Rate at 720p$0.015/sec
    Rate at 1080p$0.037/sec

    Billing is per second of output, with no subscription. The pricing page shows the live rates for every model.

    How to use the SeeDance 1.0 Pro Fast API

    1. Create an API key on the API keys page.
    2. Send a POST request to /v1/tasks with the model ID and an input object.
    3. Read the task_id from the response, then check the task until it is finished, or pass a callback URL and wait for the result.

    The same /v1/tasks endpoint handles both request types. This request is text-to-video.

    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-fast",
        "input": {
          "prompt": "A locked-off studio shot of a glass sculpture rotating slowly",
          "aspect_ratio": "9:16",
          "duration": 6,
          "resolution": "720p",
          "camera_fixed": true,
          "seed": 42
        }
      }'
    

    The response returns the task ID straight away:

    {
      "code": 200,
      "success": true,
      "data": {
        "task_id": "abc123def456",
        "status": "processing"
      }
    }
    

    The API returns a task_id. Poll GET /v1/tasks/<task_id> or provide a callback URL to handle completion without keeping the request open.

    curl https://api.unifically.com/v1/tasks/{task_id} \
      -H "Authorization: Bearer YOUR_API_KEY"
    

    SeeDance 1.0 Pro Fast API parameters

    ParameterTypeDefaultWhat it does
    promptstringnoneText prompt, up to 5,000 characters. Required for both text-to-video and image-to-video
    aspect_ratiostring16:91:1, 3:4, 4:3, 9:16, 16:9, or 21:9
    durationinteger4Output length in whole seconds, 4 to 12
    resolutionstring480p480p, 720p, or 1080p
    seedintegerrandomSeed for reproducibility. Leave it out for a random seed
    camera_fixedbooleanfalseWhen true, keeps the virtual camera stationary for the whole clip
    first_frame_urlstringnoneStart frame image URL for image-to-video. Formats: png, webp, jpeg, gif. Up to 30 MB, short side 300 to 6000 px, aspect ratio 0.4 to 2.5
    nsfw_checkbooleanfalseChecks the prompt and input images for NSFW content before generating. Off unless set to true

    The SeeDance 1.0 Pro Fast API documentation has the full reference.

    Image-to-video request

    Add first_frame_url to animate a source image. The prompt is still required, and last_frame_url is not supported.

    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-fast",
        "input": {
          "prompt": "The subject begins to walk forward slowly",
          "first_frame_url": "https://example.com/standing.jpg",
          "duration": 5,
          "resolution": "720p"
        }
      }'
    

    Key features of SeeDance 1.0 Pro Fast

    Text-to-video and image-to-video

    Start with only a prompt or add a first frame. The two modes cover open-ended drafts and shots that need an existing visual anchor.

    4–12 second videos

    Set any whole-second duration from 4 through 12, rather than choosing from only a few fixed presets.

    Output up to 1080p

    Use 480p or 720p for iteration and 1080p when a selected result needs a higher-resolution delivery file.

    Six prompt-led formats

    Text-to-video covers square, vertical, landscape, and ultrawide layouts without requiring a source image.

    What SeeDance 1.0 Pro Fast is best for

    • Fast prompt exploration

      Generate several short concepts before moving a selected direction into a more controlled workflow.

    • Social video variants

      Create 9:16, 1:1, and 16:9 versions for different placements from related prompts.

    • Single-image animation

      Add motion to an approved image when controlling an exact end frame is unnecessary.

    • Draft-to-delivery iteration

      Evaluate motion at a lower resolution, then rerun a selected setup at 1080p.

    • Stable viewpoint shots

      Use camera_fixed for demonstrations or compositions where large camera motion would be distracting.

    SeeDance 1.0 Pro Fast use cases

    Draft multiple ad concepts from text at 480p, keeping each prompt and seed in an evaluation sheet, then rerun the chosen setup at 1080p. Animate a product or character image into a short clip without defining a separate closing frame. Produce platform-specific versions by changing the text-to-video aspect ratio while holding the rest of the request constant. For comparisons, do not mix a new prompt, duration, and resolution in the same test: controlled changes produce evidence that is easier to reuse in a model review or integration decision.

    Preguntas frecuentes

    Preguntas frecuentes sobre SeeDance 1.0 Pro Fast

    SeeDance 1.0 Pro Fast is the speed-oriented route in ByteDance's SeeDance 1.0 family. It supports text-to-video and first-frame image-to-video on UnificAlly.

    With an account, the API keys page issues the key you send as Authorization: Bearer <key>. Point requests at https://api.unifically.com/v1/tasks with bytedance/seedance-1.0-pro-fast; the first call needs a prompt even when you animate an image.

    Give bytedance/seedance-1.0-pro-fast an input with a prompt (up to 5,000 characters), a duration of 4 to 12 seconds, a resolution of 480p, 720p or 1080p, and optionally camera_fixed and seed. Add first_frame_url for image-to-video; last_frame_url is not supported. Poll GET /v1/tasks/<task_id> or provide a callback URL to pick up the video.

    No. Its image-to-video mode accepts a first frame, but it does not expose last-frame interpolation. Use SeeDance 1.0 Pro when the closing frame must be supplied.

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

    Text-to-video supports 1:1, 3:4, 4:3, 9:16, 16:9, and 21:9. Image-to-video follows the supplied first frame.

    Yes. Set camera_fixed to true for a more stable viewpoint when the scene should rely on subject motion.

    The API accepts an integer seed. Record it with the prompt and other parameters when comparing settings or regenerating a selected setup.