Skip to main content
POST
/
v1
/
tasks
curl -X POST https://qingbo.dev/v1/tasks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0",
    "action": "generate",
    "prompt": "一只柴犬穿宇航服在月球上行走，电影感光影",
    "aspect_ratio": "16:9",
    "duration": 5,
    "resolution": "1080p",
    "watermark": false,
    "extra": {
      "generate_audio": true
    }
  }'

{
  "task_id": "task-wave1775285160b950328499",
  "model": "seedance-2.0",
  "action": "generate",
  "status": "queued",
  "created_at": 1775285160040,
  "progress": 0
}
Async processing mode — submit a request and receive a task ID. Use the Query Task Status endpoint to retrieve generation results. Video links are valid for 24 hours — please download promptly. 
curl -X POST https://qingbo.dev/v1/tasks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0",
    "action": "generate",
    "prompt": "一只柴犬穿宇航服在月球上行走，电影感光影",
    "aspect_ratio": "16:9",
    "duration": 5,
    "resolution": "1080p",
    "watermark": false,
    "extra": {
      "generate_audio": true
    }
  }'

{
  "task_id": "task-wave1775285160b950328499",
  "model": "seedance-2.0",
  "action": "generate",
  "status": "queued",
  "created_at": 1775285160040,
  "progress": 0
}

​
Available Models

Model IDDescription
seedance-2.0Seedance 2.0, supports native audio, first/last frame, multimodal reference, up to 2K
seedance-2.0-fastSeedance 2.0 fast version
seedance-1.5-proSeedance 1.5 professional version
seedance-1.0-pro-fastSeedance 1.0 fast version
seedance-1.0-pro-qualitySeedance 1.0 high-quality version

​
Common Parameters

​
model
string
required
Model ID
​
action
string
default:"generate"
Operation type
​
prompt
string
required
Video description text, supports Chinese and English. 2.0 supports referencing assets in the prompt with @image1, @video1, @audio1
​
aspect_ratio
string
default:"16:9"
Aspect ratio: 16:9, 9:16, 4:3, 3:4, 21:9, 1:1
​
duration
integer
default:"5"
Video duration (seconds), range depends on the model (see model tabs)
​
resolution
string
Output resolution, available values depend on the model (see model tabs)
​
image_urls
string[]
Reference image URL array, quantity limits depend on the model version
​
first_frame
string
First frame image URL
​
last_frame
string
Last frame image URL (requires first_frame)
​
seed
integer
default:"-1"
Random seed, range [-1, 2^32-1]. -1 for random. Same seed + same request produces similar results
​
watermark
boolean
default:"false"
Whether to add a watermark
​
callback_url
string
Webhook callback URL
​
extra
object
Additional parameters passed through to the upstream model (see model tabs)

​
Model-Specific Parameters

duration range: 4-15 secondsresolution options: 480p, 720p, 1080p, 2Ksize additional support: adaptive (auto-match input asset aspect ratio)Input limits: Up to 9 images (up to 30MB), up to 3 videos (up to 50MB, total duration up to 15s), up to 3 audio files (MP3/WAV, up to 15MB, total duration up to 15s)
​
extra.generate_audio
boolean
default:"false"
Native audio-video joint generation, supports dialogue lip-sync (8+ languages)
​
extra.return_last_frame
boolean
default:"false"
Return last frame image URL, useful for video clip continuation
​
extra.tools
array
Model tools, supports web search
"extra": { "tools": [{"type": "web_search"}] }
​
extra.service_tier
string
Service tier, controls the trade-off between latency and cost
When using first/last frame mode (first_frame + last_frame) in 2.0, video_urls and audio_urls cannot be passed simultaneously.

​
Generation Modes

ModeParametersSupported Versions
Text-to-videopromptAll
Image-to-videoprompt + image_urls (1 image) or first_frameAll
First/Last framefirst_frame + last_frame1.0-quality / 1.5 / 2.0
Multimodal referenceimage_urls + video_urls + audio_urls2.0 only

​
Response

​
task_id
string
Task ID
​
model
string
Model ID used
​
action
string
Operation type
​
status
string
Task status: queued
​
created_at
integer
Creation time (Unix millisecond timestamp)
​
progress
integer
Progress, initially 0