Skip to main content
POST
/
v1
/
videos
/
generations
curl --request POST \
  --url https://api.apimart.ai/v1/videos/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "wan2.5-preview",
    "prompt": "Sunset coastal highway, cinematic shot",
    "size": "16:9",
    "resolution": "720p",
    "duration": 5
  }'

{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01J9HA7JPQ9A0Z6JZ3V8M9W6PZ"
    }
  ]
}

Documentation Index

Fetch the complete documentation index at: https://docs.apimart.ai/llms.txt

Use this file to discover all available pages before exploring further.

curl --request POST \
  --url https://api.apimart.ai/v1/videos/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "wan2.5-preview",
    "prompt": "Sunset coastal highway, cinematic shot",
    "size": "16:9",
    "resolution": "720p",
    "duration": 5
  }'

{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01J9HA7JPQ9A0Z6JZ3V8M9W6PZ"
    }
  ]
}

​
Authentication

​
Authorization
string
required
All endpoints require Bearer Token authenticationGet API Key:Visit the API Key Management Page to get your API KeyAdd to request header:
Authorization: Bearer YOUR_API_KEY

​
Request Parameters

​
model
string
required
Video generation model name, fixed as wan2.5-preview
​
prompt
string
Video content descriptionRequired for text-to-video (without image_urls), optional but recommended for image-to-videoDescribe scenes, actions, styles in detail for better resultsExample: "Sunset coastal highway, cinematic shot"
​
image_urls
array<string>
Reference image URL array (only 1 image supported)Required for image-to-video mode, supports publicly accessible image URLs or Base64 encoding (data:image/png;base64,...)Example: ["https://example.com/image.jpg"]
The system automatically selects text-to-video or image-to-video mode based on whether image_urls is included. Do not pass image_urls for text-to-video mode.
​
negative_prompt
string
Negative prompt, describes unwanted contentMaximum 500 charactersExample: "blurry, low quality, distorted"
​
resolution
string
default:"720p"
Video resolutionOptions:
  • 480p - SD, supports size: 16:9, 9:16, 1:1
  • 720p - HD (default), supports size: 16:9, 9:16, 1:1, 4:3, 3:4
  • 1080p - FHD, supports size: 16:9, 9:16, 1:1, 4:3, 3:4
Default: 720p
Resolution directly affects pricing: 1080p > 720p > 480p.
480p only supports 16:9, 9:16, 1:1 ratios. Passing 4:3 or 3:4 will result in an error.
​
duration
integer
default:"5"
Video duration (seconds)Only supports: 5 or 10 secondsDefault: 5
​
size
string
default:"16:9"
Aspect ratio, only effective for text-to-video (without image_urls)Options depend on resolution:480p:
  • 16:9 - Landscape (default)
  • 9:16 - Portrait
  • 1:1 - Square
720p / 1080p:
  • 16:9 - Landscape (default)
  • 9:16 - Portrait
  • 1:1 - Square
  • 4:3 - Landscape
  • 3:4 - Portrait
Default: 16:9
For image-to-video, the aspect ratio is determined by the input image. Do not pass size, otherwise an error will be returned.
​
seed
integer
Random seed (≥0), specifying the same seed can reproduce similar resultsExample: 12345
​
prompt_extend
boolean
default:"true"
Whether to enable smart prompt rewritingSignificantly improves results for shorter prompts, but increases processing timeDefault: true
​
audio
boolean
default:"true"
Whether to automatically add audioWhen enabled, the system will generate matching audio for the videoDefault: true
This model only supports audio=true. Setting to false for silent video is not supported.
​
audio_url
string
Custom audio URL (wav/mp3, 3-30 seconds, ≤ 15MB)If the audio is longer than the video duration, it will be automatically trimmed; if shorter, the remaining part will be silent
Audio file requirements:
  • Format: wav, mp3
  • Duration: 3-30 seconds
  • Size: ≤ 15MB
​
watermark
boolean
default:"false"
Whether to add an “AI Generated” watermark (bottom right)Default: false

​
Resolution and Aspect Ratio Combinations

size and resolution combinations map to upstream pixel dimensions (only effective for text-to-video):
Aspect RatioDescription480p Size720p Size1080p Size
16:9Landscape (default)832×4801280×7201920×1080
9:16Portrait480×832720×12801080×1920
1:1Square624×624960×9601440×1440
4:3Landscape-1088×8321632×1248
3:4Portrait-832×10881248×1632
480p only supports 16:9, 9:16, 1:1 ratios. Passing 4:3 or 3:4 will result in an error. 720p and 1080p support all 5 ratios.

​
Response

​
code
integer
Response status code, 200 on success
​
data
array
Response data array

​
Use Cases

​
Case 1: Text-to-Video (Simple)

{
  "model": "wan2.5-preview",
  "prompt": "Sunset coastal highway, cinematic shot"
}

​
Case 2: Text-to-Video (Full Parameters)

{
  "model": "wan2.5-preview",
  "prompt": "City night scene, neon lights and rain-soaked streets",
  "negative_prompt": "blurry, low quality, distorted",
  "size": "16:9",
  "resolution": "720p",
  "duration": 5,
  "seed": 12345,
  "prompt_extend": true,
  "audio": true,
  "watermark": false
}

​
Case 3: Image-to-Video

{
  "model": "wan2.5-preview",
  "prompt": "Cat running on the grass",
  "image_urls": ["https://example.com/cat.jpg"],
  "resolution": "480p",
  "duration": 5
}

​
Case 4: Image-to-Video (Base64 Image)

{
  "model": "wan2.5-preview",
  "prompt": "Make the cat stand up and walk",
  "image_urls": ["data:image/png;base64,iVBORw0KGgo..."],
  "duration": 5
}

​
Case 5: Custom Audio

{
  "model": "wan2.5-preview",
  "prompt": "Person dancing to the music",
  "image_urls": ["https://example.com/dancer.jpg"],
  "audio_url": "https://example.com/music.mp3",
  "resolution": "720p",
  "duration": 10
}

​
Mode Description

​
Text-to-Video

  • prompt parameter is required
  • Do not pass image_urls
  • Use size to specify aspect ratio

​
Image-to-Video

  • image_urls parameter is required (only 1 image supported)
  • prompt is optional, used to describe expected actions
  • Aspect ratio is determined by the input image, do not pass size
The system automatically selects the mode based on whether image_urls is included
Query Task ResultsVideo generation is an asynchronous task that returns a task_id upon submission. Use the Get Task Status endpoint to query generation progress and results.