Video Generation API

Create and retrieve video generation tasks with Seedance, Doubao, or OpenAI video protocols.

Seedance and Doubao channels use the general task protocol. Models connected through the OpenAI video protocol use the /v1/videos endpoints. Supported reference asset types vary by model and channel.

Base URL

https://api.qiuqiutoken.net/v1

Seedance / Doubao task protocol

Create a task

POST/v1/video/generations

https://api.qiuqiutoken.net/v1/video/generations

Field	Required	Description
`model`	Required	Video model name, such as `doubao-seedance-2-0-260128`.
`prompt`	Required	Video content, shot, and audio instructions.
`seconds`	Optional	A string specifying the video duration, such as `"5"`.
`metadata.content`	Optional	Reference images, video, or audio. Use only with models and channels that support the corresponding reference asset.
`metadata.resolution`	Optional	Output resolution, such as `720p`.
`metadata.ratio`	Optional	Aspect ratio, such as `16:9`.
`metadata.generate_audio`	Optional	Whether to generate audio.
`metadata.watermark`	Optional	Whether to add a watermark.

Text-to-video

curl -X POST "https://api.qiuqiutoken.net/v1/video/generations" -H "Authorization: Bearer sk-XXXX" -H "Content-Type: application/json" -d '{
  "model": "doubao-seedance-2-0-260128",
  "prompt": "A woman walking through Tokyo streets, cinematic camera work",
  "seconds": "5",
  "metadata": {
    "resolution": "720p",
    "ratio": "16:9",
    "generate_audio": true,
    "watermark": false
  }
}'

Use reference assets

Place reference assets in metadata.content. Keep the text prompt in the top-level prompt field.

curl -X POST "https://api.qiuqiutoken.net/v1/video/generations" -H "Authorization: Bearer sk-XXXX" -H "Content-Type: application/json" -d '{
  "model": "doubao-seedance-2-0-260128",
  "prompt": "Use the composition, product appearance, and background music from the reference assets to create a first-person fruit tea advertisement",
  "seconds": "11",
  "metadata": {
    "content": [
      {
        "type": "image_url",
        "image_url": {
          "url": "https://ark-project.tos-cn-beijing.volces.com/doc_image/r2v_tea_pic1.jpg"
        },
        "role": "reference_image"
      },
      {
        "type": "image_url",
        "image_url": {
          "url": "https://ark-project.tos-cn-beijing.volces.com/doc_image/r2v_tea_pic2.jpg"
        },
        "role": "reference_image"
      },
      {
        "type": "video_url",
        "video_url": {
          "url": "https://ark-project.tos-cn-beijing.volces.com/doc_video/r2v_tea_video1.mp4"
        },
        "role": "reference_video"
      },
      {
        "type": "audio_url",
        "audio_url": {
          "url": "https://ark-project.tos-cn-beijing.volces.com/doc_audio/r2v_tea_audio1.mp3"
        },
        "role": "reference_audio"
      }
    ],
    "resolution": "720p",
    "ratio": "16:9",
    "generate_audio": true,
    "watermark": false
  }
}'

Retrieve a task

GET/v1/video/generations/{task_id}

https://api.qiuqiutoken.net/v1/video/generations/{task_id}

curl -X GET "https://api.qiuqiutoken.net/v1/video/generations/{task_id}" -H "Authorization: Bearer sk-XXXX"

Field	Description
`code`	Request result code.
`message`	Request result message.
`data`	Task data. After completion, the video URL is available at `data.result_url`.

OpenAI / Sora protocol

Use these endpoints with models and channels that support the OpenAI video protocol.

Create a task

POST/v1/videos

https://api.qiuqiutoken.net/v1/videos

curl -X POST "https://api.qiuqiutoken.net/v1/videos" -H "Authorization: Bearer sk-XXXX" -F "model=sora-2" -F "prompt=A calico cat playing a piano on stage"

Retrieve a task

GET/v1/videos/{task_id}

https://api.qiuqiutoken.net/v1/videos/{task_id}

Download video content

GET/v1/videos/{task_id}/content

https://api.qiuqiutoken.net/v1/videos/{task_id}/content

For complete OpenAI protocol fields and response definitions, see the OpenAI Video API Reference.