Kling 3.0 Standard

Identifier

Use the model identifier kling-v3-standard in the model field when calling /asset/start.

Capabilities

Property	Value
Reference inputs	0–2
Input types	`image` + `image`
Aspect ratios	`16:9`, `9:16`, `1:1` (default `9:16`)
Durations (seconds)	`3`, `4`, `5`, `6`, `7`, `8`, `9`, `10`, `11`, `12`, `13`, `14`, `15` (default `5`)
Elements	up to 4, 2–4 images each
Audio	yes
Max prompt length	2500 chars

Inputs

This model takes named media inputs. Pass each one as a top-level field in the request body. Each field accepts exactly one of url, asset_id, or avatar_id.

Field	Type	Required	Description
`first_frame`	image	no	Image used as the first frame of the generated video. Omit for pure text-to-video.
`last_frame`	image	no	Image used as the last frame of the generated video. Requires first_frame — generates a transition from the first frame to this one.

url — a direct, publicly accessible file URL
asset_id — the ID of an existing asset in your Hoox space (e.g. a previous generation)
avatar_id — the ID of an avatar defined in your Hoox space

A positional references array is also accepted, mapped onto the slots in the order above.

Elements

Elements are named characters or objects you inject in the generation, each described by several images of the same subject. You then refer to them in the prompt with @name, so the model knows exactly which subject a sentence is about. Pass up to 4 of them in the elements array:

Field	Type	Required	Description
`name`	string	yes	Alias used in the prompt as `@name`. Letters, digits and underscores only (max 32 chars).
`images`	array	yes	2 to 4 images of the same subject — frontal view first, then other angles. Each item takes one of `url`, `asset_id`, or `avatar_id`.

Rules enforced by the API:

Each element needs at least 2 images.
The prompt must reference every element as @name — otherwise the element would be ignored and the generation charged for nothing.
Names must be unique, and cannot be refN or elementN (reserved for the internal rewriting of prompt references).
first_frame is required whenever you use elements.

Example with elements

{
  "type": "video",
  "model": "kling-v3-standard",
  "prompt": "The woman from @hero picks up the mug and smiles at the camera",
  "first_frame": {
    "url": "https://example.com/first-frame.jpg"
  },
  "elements": [
    {
      "name": "hero",
      "images": [
        {
          "url": "https://example.com/hero-frontal.jpg"
        },
        {
          "url": "https://example.com/hero-side.jpg"
        }
      ]
    }
  ],
  "duration": 5
}

Custom settings

These keys go inside the model_settings object of the request body.

Key	Type	Default	Affects cost
`generateAudio`	boolean	`true`	yes

Pricing

Default parameters (5s): ~13 credits / generation. For exact pricing with your own parameters, call /asset/pricing with the model and any parameters that affect cost.

Example request

{
  "type": "video",
  "model": "kling-v3-standard",
  "prompt": "Describe what you want to generate",
  "first_frame": {
    "url": "https://example.com/reference.jpg"
  },
  "last_frame": {
    "url": "https://example.com/reference.jpg"
  },
  "aspect_ratio": "9:16",
  "duration": 5
}

See /asset/start for the full request/response reference, error codes, and code samples in cURL / Python / JavaScript.

Script

Generation

Export

Avatar

Voices

Folder

Assets

Identifier

Capabilities

Inputs

Elements

Example with elements

Custom settings

Pricing

Example request

​Identifier

​Capabilities

​Inputs

​Elements

​Example with elements

​Custom settings

​Pricing

​Example request

Identifier

Capabilities

Inputs

Elements

Example with elements

Custom settings

Pricing

Example request