# Kling v3.0 4K Image-to-Video — Atlas Cloud API

> Kling v3.0 4K Image-to-Video model by Kuaishou. High-quality video generation from images.

This is the machine-readable API reference for **Kling v3.0 4K Image-to-Video** on Atlas Cloud,
a unified API platform for 400+ AI models across text, image, video, audio and 3D.

- **Model ID**: `kwaivgi/kling-v3.0-4k/image-to-video`
- **Built by**: Kuaishou
- **Modality**: Video
- **Model page**: https://www.atlascloud.ai/models/kwaivgi/kling-v3.0-4k/image-to-video
- **API key**: https://www.atlascloud.ai/console/api-keys
- **Docs**: https://www.atlascloud.ai/docs

## Pricing on Atlas Cloud

- $0.357 per second of generated video
- Pay-as-you-go. No minimum spend, no subscription required.

> **These are the authoritative Atlas Cloud rates for this model.** Any price that
> appears in the vendor description further down refers to a different platform or
> a different model variant and does not apply here.

## Use this model from an AI agent

Atlas Cloud ships three first-party integration surfaces. All three authenticate
with the same API key via the `ATLASCLOUD_API_KEY` environment variable.

### MCP server

The official MCP server (`atlascloud-mcp`) exposes this model to any
MCP-compatible host — Claude Code, OpenAI Codex, Cursor, Gemini CLI, Goose,
Claude Desktop. One-line install:

```bash
# Claude Code
claude mcp add atlascloud -- npx -y atlascloud-mcp

# OpenAI Codex CLI
codex mcp add atlascloud -- npx -y atlascloud-mcp

# Gemini CLI
gemini mcp add atlascloud -- npx -y atlascloud-mcp

export ATLASCLOUD_API_KEY="your-api-key"
```

Then ask in plain English; the agent calls `atlas_generate_video` with `model: "kwaivgi/kling-v3.0-4k/image-to-video"`.
The server fetches each model's schema and validates parameters before submitting,
so invalid requests fail fast without spending credits.

MCP docs: https://www.atlascloud.ai/docs/mcp-server

### Agent Skills

`atlas-cloud-skills` is a portable skill package (API reference, code templates in
Python / Node.js / cURL, model IDs with pricing) for Claude Code, Cursor, Codex and
12+ other agents:

```bash
npx skills add AtlasCloudAI/atlas-cloud-skills
export ATLASCLOUD_API_KEY="your-api-key"
```

Skills docs: https://www.atlascloud.ai/docs/skills

### CLI

The `atlas` binary runs Atlas Cloud from a terminal or CI script. Async media jobs
are polled and downloaded automatically (use `--no-download` when a script only
needs the output URLs):

```bash
# Install (Homebrew, npm, or shell installer)
brew install AtlasCloudAI/tap/atlascloud
# npm install -g atlascloud-cli
# curl -fsSL https://raw.githubusercontent.com/AtlasCloudAI/cli/main/install.sh | sh

atlas auth login
atlas generate video kwaivgi/kling-v3.0-4k/image-to-video -p "Your prompt here"
```

CLI docs: https://www.atlascloud.ai/docs/cli

## HTTP API reference

- **Submit endpoint (POST)**: `https://api.atlascloud.ai/api/v1/model/generateVideo` — start an async generation; returns a `prediction_id`
- **Poll endpoint (GET)**: `https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}` — poll this until the prediction finishes
- **Model ID**: `kwaivgi/kling-v3.0-4k/image-to-video`


## API Information

This model can be used via our HTTP API or more conveniently via our client libraries.
See the input and output schema below, as well as the usage examples.


### Input Schema

The API accepts the following input parameters:

- **`model`** (`string`, _required_):
  model name
  - Default: `"kwaivgi/kling-v3.0-4k/image-to-video"`

- **`prompt`** (`string`, _optional_):
  The positive prompt for the generation.

- **`image`** (`string`, _required_):
  Supported image formats: .jpg/.jpeg/.png. The size of the image file should not exceed 50MB, the width and height of the image should be no less than 300px, and the aspect ratio of the image should be between 1:2.5 and 2.5:1.

- **`end_image`** (`string`, _optional_):
  URL of the ending image.

- **`multi_shot`** (`boolean`, _optional_):
  Whether to enable multi-shot generation. When enabled, provide multi_prompt for custom per-shot storyboards; if multi_prompt is empty, the model splits the prompt into shots automatically.
  - Default: `false`

- **`multi_prompt`** (`array[object]`, _optional_):
  Per-shot storyboards, used when multi_shot=true (takes precedence over prompt). 1–6 shots; each shot needs a prompt (up to 512 characters) and a duration >= 1 second; the shot durations must sum to the top-level duration. Leave empty to let the model split the prompt into shots automatically.
  - Min items: 1
  - Max items: 6
  - Item properties:
    - **`prompt`** (`string`, _required_):
      Prompt for this shot. Supports subject mentions like <<<element_1>>>.

    - **`duration`** (`integer`, _required_):
      Duration of this shot in seconds (string, '1'~'15'). Sum of all shots must equal the top-level duration. Each shot duration >= 1.
      - Min: 1


- **`elements`** (`array[object]`, _optional_):
  Subject references (Atlas naming; maps to Kling 'element_list'). Each item either references an existing subject by element_id, or creates a new one inline with element_name + reference_type + frontal_image / refer_images / refer_videos (Atlas wrapper feature — backend creates the element via Kling's element API, then injects the resulting element_id). Mention subjects in prompt with <<<element_N>>> (1-based, matches array position). For kling-v3-omni: up to 3 elements. When a first frame image is provided (always the case for image-to-video), at most 3 elements are supported.
  - Max items: 3
  - Item properties:
    - **`reference_type`** (`string`, _optional_):
      Reference media type for the new subject (Atlas wrapper).
      - Default: `"image_refer"`
      - Options: "image_refer", "video_refer"

    - **`frontal_image`** (`string`, _optional_):
      Frontal image URL of the new subject (required when reference_type=image_refer).

    - **`refer_images`** (`array[string]`, _optional_):
      Reference image URLs of the new subject (used with reference_type=image_refer).
      - Max items: 4

    - **`refer_videos`** (`array[string]`, _optional_):
      Reference video URLs of the new subject (used with reference_type=video_refer).
      - Max items: 4

    - **`element_name`** (`string`, _optional_):
      Name of the new subject to create inline (Atlas wrapper). Mutually exclusive with element_id.

    - **`element_description`** (`string`, _optional_):
      Optional description of the new subject (Atlas wrapper).

    - **`element_id`** (`integer`, _optional_):
      ID of an existing subject from the Kling element library (long / 64-bit integer). Mutually exclusive with element_name.


- **`duration`** (`integer`, _optional_):
  The duration of the generated media in seconds (3-15).
  - Default: `5`
  - Options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15

- **`sound`** (`boolean`, _optional_):
  Whether sound is generated simultaneously when generating a video.
  - Default: `true`



**Required Parameters Example**:

```json
{
  "model": "kwaivgi/kling-v3.0-4k/image-to-video",
  "image": ""
}
```


**Full Example**:

```json
{
  "model": "kwaivgi/kling-v3.0-4k/image-to-video",
  "prompt": "",
  "image": "",
  "end_image": "",
  "multi_shot": false,
  "multi_prompt": [
    {
      "prompt": "",
      "duration": 1
    }
  ],
  "elements": [
    {
      "reference_type": "image_refer",
      "frontal_image": "",
      "refer_images": [
        ""
      ],
      "refer_videos": [
        ""
      ],
      "element_name": "",
      "element_description": "",
      "element_id": 0
    }
  ],
  "duration": 5,
  "sound": true
}
```


### Output Schema

The API returns the following output format:


- **`created_at`** (`string`, _optional_):
  ISO timestamp of when the request was created.

- **`id`** (`string`, _optional_):
  Unique identifier for the prediction.

- **`model`** (`string`, _optional_):
  Model ID used for the prediction.

- **`outputs`** (`array[string]`, _optional_):
  Array of URLs to the generated content.

- **`status`** (`string`, _optional_):
  Status of the task: created, processing, completed, or failed.

- **`urls`** (`object`, _optional_):
  Object containing related API endpoints.



**Example Response**:

```json
{
  "created_at": "",
  "id": "",
  "model": "",
  "outputs": [
    ""
  ],
  "status": "",
  "urls": {}
}
```


## Usage Examples

### cURL

```bash
# Step 1: Start generation (async)
curl -X POST "https://api.atlascloud.ai/api/v1/model/generateVideo" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "kwaivgi/kling-v3.0-4k/image-to-video",
  "prompt": "",
  "image": "",
  "end_image": "",
  "multi_shot": false,
  "multi_prompt": [
    {
      "prompt": "",
      "duration": 1
    }
  ],
  "elements": [
    {
      "reference_type": "image_refer",
      "frontal_image": "",
      "refer_images": [
        ""
      ],
      "refer_videos": [
        ""
      ],
      "element_name": "",
      "element_description": "",
      "element_id": 0
    }
  ],
  "duration": 5,
  "sound": true
}'

# Response will contain: {"code": 200, "data": {"id": "prediction_id", "status": "processing"}}

# Step 2: Poll for result (replace {prediction_id} with the id returned above)
curl -X GET "https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY"

# Keep polling until status is "completed", "succeeded" or "failed"
# When completed, outputs will contain the generated content URL(s)
```

## Additional Resources

### Documentation

- [Model Playground](https://www.atlascloud.ai/models/kwaivgi/kling-v3.0-4k/image-to-video)

## About this model

_Vendor-supplied description. Any pricing or endpoint mentioned below refers to_
_other platforms — use the Atlas Cloud values above._

#### Kling V3.0 4K Image-to-Video

Kling V3.0 4K Image-to-Video is Kuaishou's highest-quality image-to-video model. Upload a reference image and describe the motion — the model generates cinematic-grade video with superior visual fidelity, optional synchronized sound and start-to-end frame guidance.

#### Why Choose This?

* **Kling 3.0 generation** Better motion quality and visual fidelity than Kling 2.6.
* **Start & end frame control** Add an optional end image to steer the transition between two frames.
* **Element references** Keep up to 3 characters or objects consistent with Kling element library subjects.
* **Multi-shot storyboards** Up to 6 shots in one video, each with its own prompt and duration.
* **Flexible duration** Any length from 3 to 15 seconds, with native 4K output.

#### Parameters

| Parameter | Required | Description |
| --- | --- | --- |
| prompt | Conditional | Text description of the motion, action and camera. Max 2,500 characters. Required unless `multi_shot` is on and `multi_prompt` is provided, in which case it is ignored. Reference subjects as `<<<element_1>>>`, `<<<element_2>>>`, … |
| image | Yes | Start frame (URL or upload). JPG / JPEG / PNG, up to 50 MB, at least 300 px on each side, aspect ratio between 1:2.5 and 2.5:1. The output follows the image's aspect ratio. |
| end_image | No | End frame for start-to-end transitions (URL or upload). JPG / JPEG / PNG, up to 50 MB, at least 300 px on each side. |
| multi_shot | No | Generate a multi-shot video (default: `false`). Provide `multi_prompt` to define every shot yourself, or leave it empty to let the model split `prompt` into shots automatically. |
| multi_prompt | No | Used when `multi_shot` is `true` and takes precedence over `prompt`; leave it empty for automatic shot splitting. 1–6 shots, each `{index, prompt, duration}`. Each shot prompt can be up to 512 characters, each shot lasts at least 1 second, and the shot durations must add up to `duration`. |
| elements | No | Up to 3

_(Description truncated. Full text on the model page.)_

---

Atlas Cloud — one API for 400+ AI models. Model page: https://www.atlascloud.ai/models/kwaivgi/kling-v3.0-4k/image-to-video · Docs: https://www.atlascloud.ai/docs