seedance-2-0 series

seedance-2-0 Series Video API Guide

Basic Information

Item	Description
Model name	`seedance-2-0`, `seedance-2-0-fast`, `seedance-2-0-nsfw`
Model ID	`seedance-2-0`, `seedance-2-0-fast`, `seedance-2-0-nsfw`
Model type	Video generation
Base URL	`https://api-platform.ope.ai`
Create task	`POST /v1/videos`
Query task	`GET /v1/videos/{task_id}`
Download result	`GET /v1/videos/{task_id}/content`

Authentication:

Authorization: Bearer YOUR_API_KEY

Prepare these variables before testing:

BASE_URL="https://api-platform.ope.ai"
API_KEY="YOUR_API_KEY"

For pricing, see Model Pricing.

1. Request Format

Use POST /v1/videos. Put text instructions in the top-level prompt, and put image, video, or audio references in input_reference.content.

You can pass output dimensions in either of these ways:

Send size directly, for example 1280x720
Send resolution + ratio, for example resolution: "720p" + ratio: "16:9", instead of size

{
  "model": "seedance-2-0-fast",
  "prompt": "Write the text prompt here",
  "size": "1280x720",
  "seconds": "4",
  "input_reference": {
    "content": [
      {
        "type": "image_url",
        "role": "reference_image",
        "image_url": {
          "url": "https://example.com/reference.png"
        }
      }
    ]
  }
}

2. Fields

Field	Required	Description
`model`	Yes	Use a valid model ID, for example `seedance-2-0`
`prompt`	Yes	Video generation prompt
`size`	Conditional	Output size, for example `1280x720` or `720x1280` (the fast series does not support 1080p). Can be replaced by `resolution` + `ratio`
`resolution`	Conditional	Resolution tier. Use it together with `ratio` instead of `size`. Values: `480p`, `720p`, `1080p`
`ratio`	Conditional	Aspect ratio. Use it together with `resolution` instead of `size`. Values: `16:9`, `4:3`, `1:1`, `3:4`, `9:16`, `21:9`
`seconds`	Yes	Duration string from `4` to `15` seconds, for example `"4"` or `"5"`
`generate_audio`	No	Whether to generate audio
`seed`	No	Random seed
`input_reference.content`	No	Multimodal reference array

Use either size or resolution + ratio; do not send both.

Supported reference types:

`type`	Field	Description
`image_url`	`image_url.url`	Image URL
`video_url`	`video_url.url`	Video URL
`audio_url`	`audio_url.url`	Audio URL

Each item may include a role. Common values include:

reference_image
reference_video
reference_audio
first_frame
last_frame

3. Basic Limits

Generation duration:

seconds must be between 4 and 15

Reference and prompt limits:

Asset type	Field / Role	Limit	Duration / Specification
Reference image	`reference_image`	`9` images	Each image `< 30MB`
Reference video	`reference_video`	`3` videos	Each video `2~15s`, total duration `<= 15s`, each video `< 50MB`
Reference audio	`reference_audio`	`3` clips	Each clip `2~15s`, total duration `<= 15s`, each clip `< 15MB`
Text prompt	`prompt` / `text`	`1` item	Chinese `<= 500` characters, English `<= 1000` words; over-limit prompts may lose details

Multimodal reference constraints:

reference_audio cannot be sent on its own.
If audio is used as BGM or voice/timbre reference, input_reference.content must contain at least 1 visual asset as the multimodal anchor.
The visual asset can be reference_image or reference_video.

4. Create Task Examples

4.1 Text-to-video

curl --request POST "$BASE_URL/v1/videos" \
  --header "Authorization: Bearer $API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "seedance-2-0-fast",
    "prompt": "An orange cat standing on a rainy Tokyo street, neon reflections, cinematic camera movement.",
    "size": "1280x720",
    "seconds": "4",
    "generate_audio": false
  }'

Typical response:

{
  "id": "cgt-20260529102017-sjqj9",
  "task_id": "cgt-20260529102017-sjqj9",
  "object": "video",
  "model": "seedance-2-0-fast",
  "status": "queued",
  "created_at": 1780021218
}

4.2 Image-to-video: single image reference

curl --request POST "$BASE_URL/v1/videos" \
  --header "Authorization: Bearer $API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "seedance-2-0-fast",
    "prompt": "Keep the subject composition stable, slowly push the camera forward, realistic style.",
    "size": "1280x720",
    "seconds": "5",
    "input_reference": {
      "content": [
        {
          "type": "image_url",
          "role": "first_frame",
          "image_url": {
            "url": "https://example.com/first-frame.png"
          }
        }
      ]
    }
  }'

4.3 Image-to-video: first and last frames

curl --request POST "$BASE_URL/v1/videos" \
  --header "Authorization: Bearer $API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "seedance-2-0-fast",
    "prompt": "Naturally transition from the first frame to the last frame with coherent camera movement and stable facial expression.",
    "size": "1280x720",
    "seconds": "5",
    "input_reference": {
      "content": [
        {
          "type": "image_url",
          "role": "first_frame",
          "image_url": {
            "url": "https://example.com/first-frame.png"
          }
        },
        {
          "type": "image_url",
          "role": "last_frame",
          "image_url": {
            "url": "https://example.com/last-frame.png"
          }
        }
      ]
    }
  }'

4.4 Multiple image references

curl --request POST "$BASE_URL/v1/videos" \
  --header "Authorization: Bearer $API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "seedance-2-0-fast",
    "prompt": "Generate a product showcase video from multiple reference images. Keep the subject consistent and move the camera smoothly from left to right.",
    "size": "1280x720",
    "seconds": "6",
    "input_reference": {
      "content": [
        {
          "type": "image_url",
          "role": "reference_image",
          "image_url": {
            "url": "https://example.com/reference-1.png"
          }
        },
        {
          "type": "image_url",
          "role": "reference_image",
          "image_url": {
            "url": "https://example.com/reference-2.png"
          }
        },
        {
          "type": "image_url",
          "role": "reference_image",
          "image_url": {
            "url": "https://example.com/reference-3.png"
          }
        }
      ]
    }
  }'

4.5 Multimodal references: image + video + audio

curl --request POST "$BASE_URL/v1/videos" \
  --header "Authorization: Bearer $API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "seedance-2-0-fast",
    "prompt": "First-person fruit tea commercial. Use image 1 as the opening composition, image 2 as the final composition, follow video 1 for motion, and use audio 1 for rhythm and atmosphere.",
    "size": "1280x720",
    "seconds": "8",
    "generate_audio": true,
    "input_reference": {
      "content": [
        {
          "type": "image_url",
          "role": "reference_image",
          "image_url": {
            "url": "https://example.com/tea-reference-1.jpg"
          }
        },
        {
          "type": "image_url",
          "role": "reference_image",
          "image_url": {
            "url": "https://example.com/tea-reference-2.jpg"
          }
        },
        {
          "type": "video_url",
          "role": "reference_video",
          "video_url": {
            "url": "https://example.com/tea-motion.mp4"
          }
        },
        {
          "type": "audio_url",
          "role": "reference_audio",
          "audio_url": {
            "url": "https://example.com/tea-rhythm.mp3"
          }
        }
      ]
    }
  }'

4.6 Use `resolution` + `ratio`

curl --request POST "$BASE_URL/v1/videos" \
  --header "Authorization: Bearer $API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "seedance-2-0-fast",
    "prompt": "Create a vertical city street video at night with neon lights and a slow forward camera move.",
    "resolution": "720p",
    "ratio": "9:16",
    "seconds": "4",
    "generate_audio": false
  }'

This is equivalent to size: "720x1280". When you send resolution + ratio, you do not need to send size.

5. Query a Task

curl --request GET "$BASE_URL/v1/videos/TASK_ID_HERE" \
  --header "Authorization: Bearer $API_KEY"

Typical response:

{
  "created_at": 1780021218,
  "id": "cgt-20260529102017-sjqj9",
  "model": "seedance-2-0-fast",
  "object": "video",
  "prompt": "An orange cat standing on a rainy Tokyo street, neon reflections, cinematic camera movement.",
  "size": "1280x720",
  "status": "SUCCESS"
}

Common task statuses:

QUEUED
IN_PROGRESS
SUCCESS
FAILURE

6. Download Video

curl --request GET "$BASE_URL/v1/videos/TASK_ID_HERE/content" \
  --header "Authorization: Bearer $API_KEY" \
  --output output.mp4

The endpoint returns a binary video stream after the task is complete.

7. Size Parameters

You can send size directly, or send resolution + ratio instead. The table below lists the available values for the Seedance 2.0 series.

`resolution`	`ratio`	Equivalent `size`
`480p`	`16:9`	`864x496`
`480p`	`4:3`	`752x560`
`480p`	`1:1`	`640x640`
`480p`	`3:4`	`560x752`
`480p`	`9:16`	`496x864`
`480p`	`21:9`	`992x432`
`720p`	`16:9`	`1280x720`
`720p`	`4:3`	`1112x834`
`720p`	`1:1`	`960x960`
`720p`	`3:4`	`834x1112`
`720p`	`9:16`	`720x1280`
`720p`	`21:9`	`1470x630`
`1080p`	`16:9`	`1920x1080`
`1080p`	`4:3`	`1664x1248`
`1080p`	`1:1`	`1440x1440`
`1080p`	`3:4`	`1248x1664`
`1080p`	`9:16`	`1080x1920`
`1080p`	`21:9`	`2206x946`

Note: 1080p is available for the Seedance 2.0 series. Seedance 2.0 fast does not support 1080p.

8. Minimal Test Order

Text-to-video
Single image reference
Query the task
Download video
Multimodal references