---
title: Kling VIDEO 4.0 Flash | Runware Docs
url: https://runware.ai/docs/models/klingai-video-4-0-flash
description: 720p video generation and editing from text, images, and references, with clips up to 20 seconds
---
# Kling VIDEO 4.0 Flash

![Kling VIDEO 4.0 Flash](https://assets.runware.ai/covers/klingai-video-4-0-flash.jpg)

Kling VIDEO 4.0 Flash is the 720p model of the Kling 4.0 family, generating clips of up to 20 seconds from text, a first-frame image, reference images, reference videos, and reusable elements. It combines several references in a single request to hold subjects, style, and motion, and it edits an existing video from a prompt. It suits social content, previews of longer productions, and other work where 720p output is enough.

- **Status**: live
- **ID**: `klingai:kling-video@4.0-flash`
- Try in Playground
- **Creator**: Kling AI
- **Release Date**: November 4, 2026
- **Capabilities**: Text to Video, Image to Video, Video to Video, Edit, Checkpoint, Audio

## Pricing

$0.0799 $0.094 from, per second · ~1m 31s per request

**Rates** (per second)

- 720p · No video input: `$0.0799 $0.094`
- 720p · Video input: `$0.107 $0.126`

**5 example generations**

- [video-edit](https://runware.ai/docs/models/klingai-video-4-0-flash/examples#example-fossil-archive-dance-teaser): `$0.647` (~1m 7s)
- [first-frame](https://runware.ai/docs/models/klingai-video-4-0-flash/examples#example-alpine-shelter-weather-resilience-film) · 720p: `$0.799` (~16m 42s)
- [text-to-video](https://runware.ai/docs/models/klingai-video-4-0-flash/examples#example-mushroom-farm-music-video-segment) · 720p: `$0.799` (~1m 36s)
- [elements](https://runware.ai/docs/models/klingai-video-4-0-flash/examples#example-tactile-map-workshop-b-roll) · 1464×628: `$0.806` (~1m 15s)
- [reference-to-video](https://runware.ai/docs/models/klingai-video-4-0-flash/examples#example-nocturnal-rooftop-apiary-relight) · 960×960: `$1.07` (~1m 31s)

times vary, measured end to end · [How pricing works](https://runware.ai/docs/models-api/pricing)

## Compatibility & Validation

Either provide `inputs.frameImages`, `inputs.referenceImages`, `inputs.video`, `inputs.referenceVideos`, or `inputs.elements`, or specify `width/height`.

---

When `inputs.elements.frontalImage` or `inputs.elements.images` is provided, `inputs.elements.frontalImage` and `inputs.elements.images` are required.

---

When `inputs.frameImages` is not provided, `positivePrompt` is required.

---

When `inputs.referenceImages` or `inputs.video` or `inputs.referenceVideos` or `inputs.elements` is provided, `positivePrompt` is required.

---

When `settings.preserveAudio` is `true`, `inputs.elements.voices` cannot be used.

---

When `settings.preserveAudio` is provided, `inputs.video` or `inputs.referenceVideos` is required.

---

`width` and `height` must be used together.

---

When `width` or `height` is provided, `positivePrompt` is required.

---

The following dimension combinations are supported:

| Configuration | Dimensions |
| --- | --- |
| `720p (16:9)` | `1280x720` |
| `720p (1:1)` | `960x960` |
| `720p (9:16)` | `720x1280` |
| `720p (~21:9)` | `1464x628` |

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/klingai-video-4-0-flash#request-tasktype)

- **Type**: `string`
- **Required**: true
- **Value**: `videoInference`

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/klingai-video-4-0-flash#request-taskuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

### [outputType](https://runware.ai/docs/models/klingai-video-4-0-flash#request-outputtype)

- **Type**: `string`
- **Default**: `URL`

Video output type.

**Allowed values**: `URL`

### [outputFormat](https://runware.ai/docs/models/klingai-video-4-0-flash#request-outputformat)

- **Type**: `string`
- **Default**: `MP4`

Specifies the file format of the generated output. The available values depend on the task type and the specific model's capabilities.

- \`MP4\`: Widely supported video container (H.264), recommended for general use.
- \`WEBM\`: Optimized for web delivery.
- \`MOV\`: QuickTime format, common in professional workflows (Apple ecosystem).

**Allowed values**: `MP4` `WEBM` `MOV`

### [outputQuality](https://runware.ai/docs/models/klingai-video-4-0-flash#request-outputquality)

- **Type**: `integer`
- **Min**: `20`
- **Max**: `99`
- **Default**: `95`

Compression quality of the output. Higher values preserve quality but increase file size.

### [webhookURL](https://runware.ai/docs/models/klingai-video-4-0-flash#request-webhookurl)

- **Type**: `string`
- **Format**: `uri`

Webhook URL that receives JSON responses via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

**Learn more** (1 resource):

- [Webhooks](https://runware.ai/docs/models-api/webhooks) (platform)

### [deliveryMethod](https://runware.ai/docs/models/klingai-video-4-0-flash#request-deliverymethod)

- **Type**: `string`
- **Default**: `async`

Determines how the API delivers task results.

**Allowed values**:

- `sync` Returns results directly in the API response. Over HTTP, a task still running after 120 seconds continues as an async task and comes back with status 'processing'. Poll for its results using getResponse.
- `async` Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse.

**Learn more** (1 resource):

- [Task Polling](https://runware.ai/docs/models-api/task-polling) (platform)

### [uploadEndpoint](https://runware.ai/docs/models/klingai-video-4-0-flash#request-uploadendpoint)

- **Type**: `string`
- **Format**: `uri`

Specifies a URL where the generated content will be automatically uploaded using the HTTP PUT method. The raw binary data of the media file is sent directly as the request body. For secure uploads to cloud storage, use presigned URLs that include temporary authentication credentials.

**Common use cases:**

- **Cloud storage**: Upload directly to S3 buckets, Google Cloud Storage, or Azure Blob Storage using presigned URLs.
- **CDN integration**: Upload to content delivery networks for immediate distribution.

```text
// S3 presigned URL for secure upload
https://your-bucket.s3.amazonaws.com/generated/content.mp4?X-Amz-Signature=abc123&X-Amz-Expires=3600

// Google Cloud Storage presigned URL
https://storage.googleapis.com/your-bucket/content.jpg?X-Goog-Signature=xyz789

// Custom storage endpoint
https://storage.example.com/uploads/generated-image.jpg
```

The content data will be sent as the request body to the specified URL when generation is complete.

### [safety](https://runware.ai/docs/models/klingai-video-4-0-flash#request-safety)

- **Path**: `safety.checkContent`
- **Type**: `object (2 properties)`

Content safety checking configuration for video generation.

#### [checkContent](https://runware.ai/docs/models/klingai-video-4-0-flash#request-safety-checkcontent)

- **Path**: `safety.checkContent`
- **Type**: `boolean`

Enable or disable content safety checking. Increases total generation time.

#### [mode](https://runware.ai/docs/models/klingai-video-4-0-flash#request-safety-mode)

- **Path**: `safety.mode`
- **Type**: `string`
- **Default**: `fast`

Safety checking mode for video generation.

**Allowed values**:

- `fast` Checks key frames.
- `full` Checks all frames.

### [ttl](https://runware.ai/docs/models/klingai-video-4-0-flash#request-ttl)

- **Type**: `integer`
- **Min**: `60`

Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`.

### [includeCost](https://runware.ai/docs/models/klingai-video-4-0-flash#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [numberResults](https://runware.ai/docs/models/klingai-video-4-0-flash#request-numberresults)

- **Type**: `integer`
- **Min**: `1`
- **Max**: `4`
- **Default**: `1`

Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

**Inputs**

Input resources for the task (images, audio, etc). These must be nested inside the \`inputs\` object.

### [referenceImages](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-referenceimages)

- **Path**: `inputs.referenceImages`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `10`
- **Size (each)**: `300 px or more`
- **File size (each)**: `up to 30 MB`
- **Aspect ratio (each)**: `1:2.5 to 2.5:1`

List of reference images (UUID, URL, Data URI, or Base64).

### [frameImages](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-frameimages)

- **Path**: `inputs.frameImages`
- **Type**: `array of strings or objects`
- **Min items**: `items: 1`
- **Size (each)**: `300 px or more`
- **File size (each)**: `up to 30 MB`
- **Aspect ratio (each)**: `1:2.5 to 2.5:1`

An array of frame-specific image inputs to guide video generation. Each item can be either a plain image input (UUID, URL, Data URI, or Base64) or an object that pairs an image with a target position in the video.

The `frameImages` parameter allows you to constrain specific frames within the video sequence, ensuring that particular visual content appears at designated points. Position can be specified using `frame` (named positions or frame indices) or `timestamp` (seconds), depending on model support. This is different from `referenceImages`, which provide overall visual guidance without constraining specific timeline positions.

When the `frame` parameter is omitted, automatic distribution rules apply:

- **1 image**: Used as the first frame.

**Examples**:

**Shorthand format:** When you don't need to specify a frame position, you can pass a plain image input directly.

```json
"frameImages": [
  "aac49721-1964-481a-ae78-8a4e29b91402"
]
```

**Object format:** When you need to specify a position, use an object with `image` and either `frame` or `timestamp` (model-dependent).

```json
"frameImages": [
  {
    "image": "aac49721-1964-481a-ae78-8a4e29b91402",
    "frame": "first"
  }
]
```

**Format 1: string[]**:

- **Type**: `string`

Image input (UUID, URL, Data URI, or Base64).

**Format 2: object[]**:

#### [image](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-frameimages-format-2-image)

- **Path**: `inputs.frameImages.image`
- **Type**: `string`
- **Required**: true

Image input (UUID, URL, Data URI, or Base64).

#### [frame](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-frameimages-format-2-frame)

- **Path**: `inputs.frameImages.frame`
- **Type**: `object`

Target frame position for the image. This model only supports the first frame.

**Allowed values**:

- `first` First frame of the video.
- `0` Frame index 0 (first frame).

### [video](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-video)

- **Path**: `inputs.video`
- **Type**: `string`
- **Size**: `700 px to 4553 px`
- **Area**: `up to 8.29 MP`
- **Frame rate**: `24 fps to 60 fps`
- **Duration (each)**: `3s to 20s`
- **File size**: `up to 300 MB`
- **Aspect ratio**: `1:2.5 to 2.5:1`

Video input (UUID or URL). Source video for prompt-based editing.

### [referenceVideos](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-referencevideos)

- **Path**: `inputs.referenceVideos`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `5`
- **Size (each)**: `700 px to 4553 px`
- **Area (each)**: `up to 8.29 MP`
- **Frame rate (each)**: `24 fps to 60 fps`
- **Duration (each)**: `3s to 20s`
- **File size (each)**: `up to 300 MB`
- **Aspect ratio (each)**: `1:2.5 to 2.5:1`

List of reference videos (UUID or URL). Videos inside `elements` count toward the limit of 5 input videos and toward the combined duration limit.

### [elements](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements)

- **Path**: `inputs.elements`
- **Type**: `array of objects (7 properties)`
- **Min items**: `1`
- **Max items**: `7`

Elements allow you to include reusable assets (images, videos, or voices) in your video generation. Each element is identified by an `id` and can be referenced in the prompt using `<<<element_1>>>`, `<<<element_2>>>`, etc. in order of appearance.

An element can contain:

- **Images** via `frontalImage` and optionally `images` (up to 3 additional angles).
- **Videos** via `videos` (cannot be combined with images).
- **Voices** via `voices` (can only be combined with images, not videos).

**Examples**:

**Create a new element with an image:**

```json
"positivePrompt": "A video of <<<element_1>>> walking through a futuristic city",
"inputs": {
  "elements": [
    {
      "id": "my-character-id",
      "description": "A young woman with red hair",
      "frontalImage": "c64351d5-4c59-42f7-95e1-eace013eddab",
      "tags": ["Character"]
    }
  ]
}
```

**Reuse a previously created element by ID:**

```json
"positivePrompt": "A video of <<<element_1>>> sitting in a coffee shop, reading a book",
"inputs": {
  "elements": [
    {
      "id": "my-character-id"
    }
  ]
}
```

#### [id](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-id)

- **Path**: `inputs.elements.id`
- **Type**: `string`
- **Required**: true
- **Min**: `1`
- **Max**: `20`

Unique identifier for this element. Use to create a new element or reference a previously created one.

#### [description](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-description)

- **Path**: `inputs.elements.description`
- **Type**: `string`
- **Required**: true
- **Min**: `1`
- **Max**: `100`

Description of the element.

#### [frontalImage](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-frontalimage)

- **Path**: `inputs.elements.frontalImage`
- **Type**: `string`

Frontal reference image for the element. Required when using image-based elements.

#### [images](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-images)

- **Path**: `inputs.elements.images`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `3`

Reference images for the element. Up to 3 images. Requires frontalImage.

#### [videos](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-videos)

- **Path**: `inputs.elements.videos`
- **Type**: `array of strings`
- **Min items**: `items: 1`

Reference video for the element. Cannot be combined with images or voices.

#### [voices](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-voices)

- **Path**: `inputs.elements.voices`
- **Type**: `array of strings`
- **Min items**: `items: 1`

Voice audio for the element. Can only be combined with images, not videos.

#### [tags](https://runware.ai/docs/models/klingai-video-4-0-flash#request-inputs-elements-tags)

- **Path**: `inputs.elements.tags`
- **Type**: `array of strings`
- **Min items**: `1`

Classification tags for the element.

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/klingai-video-4-0-flash#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `klingai:kling-video@4.0-flash`

Send `auto` instead of an AIR identifier to have the model router choose the model for this task. The rest of the request is unchanged, every value you send is kept, and only the parameters you leave out are filled in. See [Model Router](https://runware.ai/docs/models-api/model-router).

**Learn more** (3 resources):

- [Text To Image: Model Selection](https://runware.ai/docs/learn/text-to-image#model-selection) (learn)
- [Image Inpainting: Model Specialized Inpainting Models](https://runware.ai/docs/learn/image-inpainting#model-specialized-inpainting-models) (learn)
- [Image Outpainting: Other Critical Parameters](https://runware.ai/docs/learn/image-outpainting#other-critical-parameters) (learn)

### [positivePrompt](https://runware.ai/docs/models/klingai-video-4-0-flash#request-positiveprompt)

- **Type**: `string`
- **Required**: conditional — Required when inputs.frameImages is not provided or inputs.referenceImages and inputs.video and inputs.referenceVideos and inputs.elements are provided or width and height are provided
- **Min**: `1`
- **Max**: `26000`

Text prompt describing elements to include in the generated output. Images are referenced as `<<<image_1>>>` and videos as `<<<video_1>>>`. Images are numbered across `frameImages` then `referenceImages`, and videos across `video` then `referenceVideos`. Must satisfy `E / 26000 + N / 11000 <= 1`, where `N` counts the non-ASCII characters and `E` the rest.

**Learn more** (1 resource):

- [Prompts](https://runware.ai/docs/learn/prompts) (learn)

### [width](https://runware.ai/docs/models/klingai-video-4-0-flash#request-width)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.frameImages and inputs.referenceImages and inputs.video and inputs.referenceVideos and inputs.elements are not provided
- **Paired with**: height

Width of the generated media in pixels.

**Learn more** (2 resources):

- [Dimensions](https://runware.ai/docs/learn/dimensions) (learn)
- [Image Outpainting: Dimensions Critical For Outpainting](https://runware.ai/docs/learn/image-outpainting#dimensions-critical-for-outpainting) (learn)

### [height](https://runware.ai/docs/models/klingai-video-4-0-flash#request-height)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.frameImages and inputs.referenceImages and inputs.video and inputs.referenceVideos and inputs.elements are not provided
- **Paired with**: width

Height of the generated media in pixels.

**Learn more** (2 resources):

- [Dimensions](https://runware.ai/docs/learn/dimensions) (learn)
- [Image Outpainting: Dimensions Critical For Outpainting](https://runware.ai/docs/learn/image-outpainting#dimensions-critical-for-outpainting) (learn)

### [duration](https://runware.ai/docs/models/klingai-video-4-0-flash#request-duration)

- **Type**: `integer`
- **Min**: `3`
- **Max**: `20`
- **Default**: `10`

Length of the generated video in seconds. The total number of frames produced is determined by duration multiplied by the model's frame rate (fps).

**Settings**

Technical parameters to fine-tune the inference process. These must be nested inside the \`settings\` object.

### [preserveAudio](https://runware.ai/docs/models/klingai-video-4-0-flash#request-settings-preserveaudio)

- **Path**: `settings.preserveAudio`
- **Type**: `boolean`

Keep the audio of the input video instead of generating new audio. The audio comes from `video`, or from the first `referenceVideos` item when no `video` is sent.

## Response Parameters

### [taskType](https://runware.ai/docs/models/klingai-video-4-0-flash#response-tasktype)

- **Type**: `string`
- **Required**: true
- **Value**: `videoInference`

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/klingai-video-4-0-flash#response-taskuuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID v4 identifier echoed from the original request, used to match async responses to their tasks.

### [videoUUID](https://runware.ai/docs/models/klingai-video-4-0-flash#response-videouuid)

- **Type**: `string`
- **Required**: true
- **Format**: `UUID v4`

UUID of the output video.

### [videoURL](https://runware.ai/docs/models/klingai-video-4-0-flash#response-videourl)

- **Type**: `string`
- **Format**: `uri`

URL of the output video.

### [videoBase64Data](https://runware.ai/docs/models/klingai-video-4-0-flash#response-videobase64data)

- **Type**: `string`

Base64-encoded video data.

### [videoDataURI](https://runware.ai/docs/models/klingai-video-4-0-flash#response-videodatauri)

- **Type**: `string`
- **Format**: `uri`

Data URI of the output video.

### [seed](https://runware.ai/docs/models/klingai-video-4-0-flash#response-seed)

- **Type**: `integer`

The seed used for generation. If none was provided, shows the randomly generated seed.

### [NSFWContent](https://runware.ai/docs/models/klingai-video-4-0-flash#response-nsfwcontent)

- **Type**: `boolean`

Flag indicating if NSFW content was detected.

### [cost](https://runware.ai/docs/models/klingai-video-4-0-flash#response-cost)

- **Type**: `float`

Task cost in USD. Present when `includeCost` is set to `true` in the request.

## Examples

### text-to-video (text-to-video)

[Watch video](https://assets.runware.ai/examples/klingai-video-4-0-flash/60481910-8a30-453a-88c6-357603229939.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "e98b1702-57f9-4641-ba69-04709acdf36a",
  "model": "klingai:kling-video@4.0-flash",
  "positivePrompt": "Create a finished landscape music-video segment for the instrumental bridge of an electro-folk song. The setting is a vast commercial mushroom farm at night: long symmetrical aisles of stainless-steel growing racks filled with oyster mushrooms, damp concrete floors, faint mist, cool cyan fluorescent tubes, and isolated amber work lights. Five adult farm workers in practical cream coveralls, rubber boots, gloves, and translucent hair coverings perform precise, grounded choreography with empty blue harvest crates. Their movement feels rhythmic and physically believable rather than theatrical: sliding crates down the aisle, catching them, pivoting in unison, stepping across the wet floor, and striking the crate sides like percussion instruments.\n\nBegin with an extreme low-angle view inches above the reflective floor as one crate slides rapidly toward the lens. The camera retreats smoothly ahead of it, then rises into a fast stabilized tracking shot that reveals the workers emerging one by one from between the racks. Their actions build in complexity across the shot: first a single worker pushes the crate, then two workers intercept and redirect it, then all five join in a synchronized traveling formation. The camera arcs around them while continuing forward, creating strong parallax from the repeating racks and mushroom clusters. Loose condensation shakes from the shelving as boots stamp in rhythm; tiny droplets and pale mushroom spores catch the backlight.\n\nAt the musical peak, the workers toss the lightweight empty crates between one another in a clean cascading pattern while advancing toward camera. Use a brief controlled speed ramp as the central crate rotates through the amber light, then return to natural motion when it is caught. The camera passes through the formation, whip-pans to reverse direction, and tracks backward as the workers split symmetrically into adjacent aisles. End on a bold wide composition: the central aisle suddenly empty, the five workers visible in alternating gaps between racks, each striking a final still pose as the overhead lights switch off sequentially from the far end toward camera. The last remaining amber lamp silhouettes drifting mist and rows of oyster mushrooms.\n\nArt direction: tactile agricultural realism combined with disciplined contemporary dance, graphic repetition, cool cyan and warm amber palette, glossy wet-floor reflections, subtle 16mm film grain, crisp wardrobe continuity, cinematic contrast, energetic editorial pacing achieved within a coherent continuous-shot illusion. Preserve realistic anatomy, stable faces, consistent clothing, convincing crate weight and trajectories, and safe workplace behavior. No captions, lyrics, titles, logos, labels, watermarks, fantasy imagery, or visible branding.",
  "width": 1280,
  "height": 720
}
```

---

### first-frame (first-frame)

[Watch video](https://assets.runware.ai/examples/klingai-video-4-0-flash/42581c2e-caa9-4875-a3de-ad13848ed465.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "45800f10-f1d9-4012-bce0-b48217dcf85a",
  "model": "klingai:kling-video@4.0-flash",
  "positivePrompt": "Begin exactly from <<<image_1>>> and animate the depicted alpine shelter scene forward as one uninterrupted, photorealistic architectural shot. The camera makes a very slow, stable push toward the shelter with subtle natural handheld weight, preserving the original portrait composition, terrain, mountain geography, shelter design, proportions, cladding, roof, window, door and beacon throughout. Calm late-summer air gradually changes into dangerous mountain weather within the same continuous take: the dry grass first stirs in a rising breeze, high clouds gather and move across the peaks, sunlight cools and fades, then stronger gusts bend the grass and carry loose leaves and fine dust across the ridge. A bank of slate-gray cloud rolls naturally through the background, briefly obscuring the distant mountains. Light rain begins, visible against the dark cladding, and droplets streak across the shelter surfaces. The temperature appears to fall; rain turns progressively to sleet and then dense wind-driven snow. Frost develops along the roof edges and a thin, uneven layer of fresh snow accumulates on the roof, ground and windward side without changing the underlying architecture. As daylight darkens, the shelter window illuminates from within with a steady amber glow and the small exterior safety beacon begins a slow, realistic pulse, communicating reliable operation during severe weather. End in a convincing early-winter whiteout with swirling snow, reduced mountain visibility and the shelter still clearly readable as the stable hero subject. Maintain physically plausible weather progression, continuous motion, coherent wind direction, stable geometry and realistic snow accumulation. No cuts, no time-lapse jumps, no camera orbit, no people appearing, no structural transformation, no extra buildings, no lettering, no captions, no logos. Synchronized natural location audio only: soft ridge wind growing into forceful gusts, grass rustle, rain tapping metal, then muffled blowing snow and a faint low mechanical hum from the operating shelter; no music and no speech.",
  "width": 720,
  "height": 1280,
  "inputs": {
    "frameImages": [
      "https://assets.runware.ai/assets/inputs/0c939417-4c69-4646-8005-9f3effe01c83.jpg"
    ]
  }
}
```

---

### reference-to-video (reference-to-video)

[Watch video](https://assets.runware.ai/examples/klingai-video-4-0-flash/67ed397b-078d-4c1a-b2bc-fa0551407891.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "3a805b40-c32b-48b7-8f0b-789253ab0198",
  "model": "klingai:kling-video@4.0-flash",
  "positivePrompt": "Create a polished day-for-night relight for an urban ecology documentary. Use <<<video_1>>> as the authoritative reference for the complete shot content, timing, camera push-in, framing, beekeeper identity and clothing, hive placement, hand actions, drifting bees, rooftop architecture and distant skyline. Preserve that documentary action and spatial continuity without adding, removing or relocating subjects or objects. Use <<<image_1>>> only as the lighting, color and atmosphere target: convincing indigo night, cool moonlit edge light, soft cyan ambient sky fill, deep but readable shadows, restrained warm amber illumination from distant apartment windows, subtle haze and realistic nocturnal contrast. Do not copy objects or composition from the image reference.\n\nTransform the daylight footage into a seamless late-night rooftop scene. Remove all signs of daytime illumination and bright daytime sky while retaining natural material detail in the white bee suit, weathered timber hives, concrete and metal vents. The beekeeper remains clearly readable without appearing artificially spotlighted. Moonlight should fall consistently from one off-camera direction and remain stable as the camera moves. Give reflective metal surfaces small cool highlights and let the hive interiors retain faint warm brown tones. Keep the bees visible as occasional tiny silhouettes crossing the moonlit air; do not enlarge or multiply them. Replace the distant daytime skyline illumination with believable sparse window lights and dim red rooftop aviation beacons. Maintain realistic exposure, coherent moving shadows, stable faces and hands, temporally consistent textures and natural motion blur. No fantasy glow, excessive teal-orange grading, crushed blacks, flicker, morphing, new props, signage, captions, logos or cuts. Square composition, premium natural-history documentary finish. Audio should be understated and realistic: quiet rooftop wind, faint city ambience, soft hive handling sounds and a low natural bee hum, with no music or speech.",
  "width": 960,
  "height": 960,
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/40464d43-d1a6-4fc3-a756-039bd250bea0.jpg"
    ],
    "referenceVideos": [
      "https://assets.runware.ai/assets/inputs/5ae25f68-09ac-4421-b07b-37a8c85cc0fa.mp4"
    ]
  }
}
```

---

### video-edit (video-edit)

[Watch video](https://assets.runware.ai/examples/klingai-video-4-0-flash/1705b415-c912-4ac4-94fa-a4d828de50ab.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "cd86762f-022a-4a72-809e-699397e46493",
  "model": "klingai:kling-video@4.0-flash",
  "positivePrompt": "Edit <<<video_1>>>. Replace only the plain warm-gray cyclorama behind the performer with a photorealistic natural history museum fossil archive: tall powder-coated steel specimen cabinets receding in orderly rows, several articulated dinosaur vertebrae and plaster-jacketed fossils resting on shelving, a glass-fronted preparation bench in the distant background, bone ivory and muted oxblood surfaces with small cobalt accents, warm amber practical work lights, restrained institutional 1970s design, realistic depth and scale, subtle atmospheric perspective, and background illumination matched naturally to the existing studio light. Keep the replacement entirely behind the performer and preserve clean separation around the hair, scarf, hands, and moving limbs. Keep everything else unchanged: the same performer, face, body, rust coveralls, cobalt scarf, exact choreography, gestures, pivot, low finishing stance, movement speed, timing, clip length, frame cadence, foreground studio floor, contact shadow, lighting on the performer, spatial placement, medium-wide framing, lens perspective, locked waist-height camera, continuous single take, and absence of text, captions, titles, or logos.",
  "settings": {
    "preserveAudio": true
  },
  "inputs": {
    "video": "https://assets.runware.ai/assets/inputs/8ba47614-1ff0-4d13-947f-b0722b06148f.mp4"
  }
}
```

---

### elements (elements)

[Watch video](https://assets.runware.ai/examples/klingai-video-4-0-flash/8e78ba98-f1ba-450a-b3b4-613693477baa.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "d60ac1f8-b064-4591-b32a-85e0dc57707c",
  "model": "klingai:kling-video@4.0-flash",
  "positivePrompt": "Create a polished editorial b-roll shot inside a contemporary tactile-map fabrication workshop. <<<element_1>>> stands alone at a large illuminated inspection table, lifts a freshly molded translucent navigation panel from its cooling bed, tilts it through a band of raking light, and carefully checks the raised routes and tactile symbols with her fingertips. She then looks naturally toward the camera and says in clear American English: \"A map should make the route understandable before the first step.\" Accurate synchronized speech and restrained, confident delivery. One continuous cinematic take with a slow lateral dolly from a wide environmental composition into a medium-wide view; keep her hands, the complete panel, and the fabrication equipment readable throughout. Ultra-wide landscape framing with layered workshop depth, ivory polymer sheets, cobalt fixtures, brushed aluminum tools, and warm amber practical lights. Clean contemporary industrial design, soft directional daylight, realistic skin and materials, subtle machine movement in the background, premium documentary photography, calm and purposeful mood. Natural production audio: quiet ventilation, a gentle plastic-sheet flex, subdued equipment hum, and clear dialogue. No other people, no logos, no captions, no added titles, and no legible printed text.",
  "width": 1464,
  "height": 628,
  "inputs": {
    "elements": [
      {
        "id": "tactile-guide_7314",
        "description": "A tactile wayfinding specialist with a calm, precise manner.",
        "frontalImage": "https://assets.runware.ai/assets/inputs/cf65a20f-a09b-4f35-ae9a-a7fab955bd03.jpg",
        "images": [
          "https://assets.runware.ai/assets/inputs/284f219d-8afe-474a-a48c-7eac20e1cb5e.jpg"
        ],
        "voices": [
          "https://assets.runware.ai/assets/inputs/0871932f-a568-4f5d-9aa5-d979a58b3267.mp3"
        ]
      }
    ]
  }
}
```