Elements To Video

Version: v3.0
Last Updated: 2026-04-17
Maintained by: Video Rebirth team

Multi-element video generation API - Create dynamic videos from multiple images and text descriptions

API Endpoints

Create Task: POST /api/vdr/videos/elements2video
Query Task: GET /api/vdr/videos/elements2video/{task_id}
Authentication: Bearer Token

Create Task

Endpoint: POST /api/vdr/videos/elements2video

Request Example:

bash

Copy

curl -X POST 'https://api-gen-na.bach.art/api/vdr/videos/elements2video' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "reference_inputs": [
      {
        "type": "subject",
        "images": [
          "https://example.com/hero-front.jpg",
          "https://example.com/hero-side.jpg"
        ],
        "subject": {
          "subject_name": "Hero Character",
          "subject_desc": "Warrior in golden armor",
          "subject_style": "realistic",
          "subject_type": "character"
        }
      },
      {
        "type": "image",
        "url": "https://example.com/castle-background.jpg"
      }
    ],
    "prompt": "Hero walks toward castle, sunlight gleaming on armor, camera slowly pushes in",
    "negative_prompt": "blur, low quality, distorted",
    "duration": 6,
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "fps": 30,
    "seed": 42,
    "generate_audio": false
  }'

Success Response:

json

Copy

{
  "code": 0,
  "message": "Success",
  "data": {
    "task_id": "550e8400-e29b-41d4-a716-446655440000",
    "created_at": 1678886400
  }
}

Request Parameters

Parameter	Type	Required	Default	Description
`reference_inputs`	Array	✅ Yes	-	Reference input list, max 9 elements; supports `image` and `subject` types, see details below
`prompt`	String	✅ Yes	-	Video content description, max 10000 characters
`duration`	Number	✅ Yes	-	Video duration in seconds, range: 1-6
`resolution`	String	✅ Yes	-	Video resolution: `720p` / `1080p`
`aspect_ratio`	String	No	`"16:9"`	Aspect ratio: `"16:9"` landscape / `"9:16"` portrait / `"1:1"` square
`negative_prompt`	String	No	-	Unwanted content description, max 10000 characters
`fps`	Number	No	`30`	Frame rate: `24` / `30`
`seed`	Number	No	Random	Random seed, range 0-10000000, same seed produces similar results
`generate_audio`	Boolean	No	`false`	Whether to generate audio
`audio_prompt`	String	No	-	Audio description, max 200 characters, requires `generate_audio: true`
`callback_url`	String	No	-	Callback URL (HTTP/HTTPS), max 500 characters, notified upon completion

reference_inputs Details

Supports two input types:

Type 1: Single Image (`type: "image"`)

Use a single image as video material.

json

Copy

{
  "type": "image",
  "url": "<your image url>"
}

Parameter	Type	Required	Default	Description
`type`	String	✅ Yes	-	Fixed value `"image"`
`url`	String	✅ Yes	-	Publicly accessible HTTP/HTTPS image URL

Type 2: Subject Composition (`type: "subject"`)

Subject refers to the core object in the video that requires consistent appearance characteristics, such as a character, brand IP, or specific item. By providing 2-4 multi-angle photos of the same object, the model can maintain stable reference to the subject's appearance, clothing, and overall style during video generation, improving subject consistency.

To generate custom subjects from text descriptions, see Custom Subject Documentation.

Use 2-4 photos of the same character or object to establish subject reference for video generation.

json

Copy

{
  "type": "subject",
  "images": [
    "https://example.com/hero-front.jpg",
    "https://example.com/hero-side.jpg"
  ],
  "subject": {
    "subject_name": "Hero Character",
    "subject_desc": "Warrior in golden armor",
    "subject_style": "realistic",
    "subject_type": "character"
  }
}

Parameter	Type	Required	Default	Description
`type`	String	✅ Yes	-	Fixed value `"subject"`
`images`	Array	✅ Yes	-	List of 2-4 image URLs
`subject.subject_name`	String	✅ Yes	-	Subject name, max 50 characters
`subject.subject_desc`	String	No	-	Subject description, max 500 characters
`subject.subject_style`	String	No	`"realistic"`	Style: `realistic` / `anime` / `cartoon` / `3d` / `pixel`
`subject.subject_type`	String	No	`"character"`	Subject type: `character` / `object`

Response Parameters

Create Task Response

Success (code: 0):

json

Copy

{
  "code": 0,
  "message": "Success",
  "data": {
    "task_id": "550e8400-e29b-41d4-a716-446655440000",
    "created_at": 1678886400
  }
}

Failure (code: non-zero):

json

Copy

{
  "code": 1000,
  "message": "Specific error reason",
  "data": {}
}

Query Task

Endpoint: GET /api/vdr/videos/elements2video/{task_id}

Path Parameters:

task_id (string, required): Task ID returned from create task request

Request Example:

bash

Copy

curl -X GET 'https://api-gen-na.bach.art/api/vdr/videos/elements2video/550e8400-e29b-41d4-a716-446655440000' \
  -H 'Authorization: Bearer YOUR_API_KEY'

Response:

json

Copy

{
  "code": 0,
  "message": "Success",
  "data": {
    "task_id": "550e8400-e29b-41d4-a716-446655440000",
    "status": "TASK_SUCCEEDED",
    "video_url": "https://cdn.bach.art/videos/xxx.mp4"
  }
}

Task Status Values

Status	Description
`TASK_PENDING`	Task submitted, awaiting processing
`TASK_PROCESSING`	Task in progress
`TASK_SUCCEEDED`	Video generation succeeded
`TASK_FAILED`	Video generation failed

Completion Example:

json

Copy

{
  "code": 0,
  "data": {
    "task_id": "550e8400-xxx",
    "status": "TASK_SUCCEEDED",
    "video_url": "https://cdn.bach.art/videos/xxx.mp4"
  }
}

Text to Subject

Batch upload Asset

On this page

Elements To Video

API Endpoints

Create Task

Request Parameters

reference_inputs Details

Type 1: Single Image (type: "image")

Type 2: Subject Composition (type: "subject")

Response Parameters

Create Task Response

Query Task

Task Status Values

Type 1: Single Image (`type: "image"`)

Type 2: Subject Composition (`type: "subject"`)