---
title: Wan3.0 | Runware Docs
url: https://runware.ai/docs/models/alibaba-wan3-0
description: All-in-one multimodal video generation with native 30-second clips, large reference capacity, and precise video editing
---
# Wan3.0

![Wan3.0](https://assets.runware.ai/covers/alibaba-wan3-0.jpg)

Wan3.0 is Alibaba's higher-end all-in-one multimodal video model for longer, reference-heavy, production-oriented creation. It combines text-to-video, first- and last-frame image-to-video, reference-driven video generation, localized video editing, and temporal extension in one model, with support for image, video, audio, document, and webpage inputs. It is especially well suited to branded storytelling, product demos, explainers, music-led sequences, and other commercial workflows that need stronger character or product consistency, more impactful audiovisual motion, and tighter control over complex multi-input prompts.

- **Status**: live
- **ID**: `alibaba:wan@3.0`
- Try in Playground
- **Creator**: Alibaba
- **Release Date**: August 24, 2026
- **Capabilities**: Text to Video, Image to Video, Video to Video, Audio to Video, Edit, Extend, Checkpoint, Audio

## Pricing

$0.015 $0.05 from, per second · ~3m 12s per request

**Rates** (per second)

- 480p: `$0.015 $0.05`
- 720p: `$0.03 $0.10`
- 1080p: `$0.06 $0.20`

**5 example generations**

- [first-frame](https://runware.ai/docs/models/alibaba-wan3-0/examples#example-mycelium-packaging-factory-b-roll) · 720p · 5s: `$0.504` (~2m 8s)
- [text-to-video](https://runware.ai/docs/models/alibaba-wan3-0/examples#example-restaurant-knife-sharpening-service-reel) · 720p · 6s · audio: `$0.604` (~3m 6s)
- [url-grounded](https://runware.ai/docs/models/alibaba-wan3-0/examples#example-climbing-helmet-colorway-reveal) · 960×960 · 7s: `$0.704` (~8m 43s)
- [first-last-frame](https://runware.ai/docs/models/alibaba-wan3-0/examples#example-airport-falconry-team-introduction) · 1080p · 8s: `$1.62` (~3m 12s)
- [reference-to-video](https://runware.ai/docs/models/alibaba-wan3-0/examples#example-hospital-tube-network-documentary-teaser) · 1080p · 10s: `$2.01` (~7m 4s)

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

## Compatibility & Validation

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

---

Inside `inputs`, `frameImages` cannot be used with `referenceImages`, `referenceVideos`, or `referenceAudios`.

---

Inside `inputs`, `documents` cannot be used with `urls`.

---

When none of `inputs.frameImages`, `inputs.referenceImages`, `inputs.referenceVideos`, or `inputs.referenceAudios` are provided, `positivePrompt` is required.

---

`resolution` cannot be used with `width/height`.

---

When `resolution` is provided, `inputs.frameImages`, `inputs.referenceImages`, or `inputs.referenceVideos` is required.

---

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

---

The following dimension combinations are supported:

| Configuration | Dimensions |
| --- | --- |
| `480p (~16:9)` | `832x480` |
| `480p (~9:16)` | `480x832` |
| `480p (1:1)` | `624x624` |
| `480p (~4:3)` | `720x544` |
| `480p (~3:4)` | `544x720` |
| `720p (16:9)` | `1280x720` |
| `720p (9:16)` | `720x1280` |
| `720p (1:1)` | `960x960` |
| `720p (~4:3)` | `1104x832` |
| `720p (~3:4)` | `832x1104` |
| `1080p (16:9)` | `1920x1080` |
| `1080p (9:16)` | `1080x1920` |
| `1080p (1:1)` | `1440x1440` |
| `1080p (~4:3)` | `1648x1248` |
| `1080p (~3:4)` | `1248x1648` |

## Notes

Wan 3.0 decides what to do from the inputs you attach:

- **Text-to-Video:** Provide a prompt on its own.
- **Keyframes:** Provide `inputs.frameImages` (up to 2) to pin the first and last frame.
- **References:** Provide `inputs.referenceImages` (up to 10), `inputs.referenceVideos` (up to 5), and `inputs.referenceAudios` (up to 5) to guide subject, motion, and sound. These combine with each other, but not with `frameImages`.
- **Source material:** Provide `inputs.documents` or `inputs.urls` (one item each) to build the video from a file or a webpage. The two are mutually exclusive.

Reference assets in the prompt by array order, as `Image 1`, `Video 1`, and `Audio 1`.

**Sizing:** Keyframes, reference images and reference videos make dimensions optional. Send `resolution` (`480p`, `720p`, `1080p`) to pick the tier and inherit the source aspect ratio, or a `width`/`height` pair to pin the frame exactly. Every other request needs an explicit pair, including document and URL requests, where `resolution` is rejected even though a file is attached.

**Duration:** Any whole number of seconds from 2 to 30, or `auto` to fit the content.

> [!WARNING]
> When `inputs.referenceVideos` is attached, `duration` and the reference clip share a single 30-second budget, so a 12-second reference leaves at most 18 seconds of output.

**Prompt rewriting:** `settings.promptExtend` is on by default, so an LLM expands your prompt before generation and the same `seed` will not reproduce an earlier result. Turning it off makes runs repeatable and saves roughly 20 to 60 seconds of latency, at the cost of quality unless your prompt is already detailed.

**Audio:** `settings.audio` is on by default and generates a native soundtrack aligned to the visuals.

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/alibaba-wan3-0#request-tasktype)

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

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0#request-outputtype)

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

Video output type.

**Allowed values**: `URL`

### [outputFormat](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0#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/alibaba-wan3-0#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/alibaba-wan3-0#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/alibaba-wan3-0#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/alibaba-wan3-0#request-safety)

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

Content safety checking configuration for video generation.

#### [checkContent](https://runware.ai/docs/models/alibaba-wan3-0#request-safety-checkcontent)

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

Enable or disable content safety checking. Increases total generation time and adds a small charge on top of the generation cost.

#### [mode](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0#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/alibaba-wan3-0#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [numberResults](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0#request-inputs-referenceimages)

- **Path**: `inputs.referenceImages`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `10`
- **File size (each)**: `up to 20 MB`
- **Size (each)**: `240 px or more`
- **Aspect ratio (each)**: `1:8 to 8:1`

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

### [frameImages](https://runware.ai/docs/models/alibaba-wan3-0#request-inputs-frameimages)

- **Path**: `inputs.frameImages`
- **Type**: `array of strings or objects`
- **Min items**: `1`
- **Max items**: `2`
- **File size (each)**: `up to 20 MB`
- **Size (each)**: `240 px or more`
- **Aspect ratio (each)**: `1:8 to 8:1`

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.
- **2 images**: First and last frames.

**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"
  }
]
```

**First and last frames:** With two images, they automatically become the first and last frames of the video sequence. You can mix shorthand and object formats.

```json
"frameImages": [
  "aac49721-1964-481a-ae78-8a4e29b91402",
  {
    "image": "3ad204c3-a9de-4963-8a1a-c3911e3afafe",
    "frame": "last"
  }
]
```

**Format 1: string[]**:

- **Type**: `string`

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

**Format 2: object[]**:

#### [image](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0#request-inputs-frameimages-format-2-frame)

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

Target frame position for the image. Supports first and last frame.

**Allowed values**:

- `first` First frame of the video.
- `last` Last frame of the video.
- `0` Frame index 0 (first frame).
- `-1` Frame index -1 (last frame).

### [referenceVideos](https://runware.ai/docs/models/alibaba-wan3-0#request-inputs-referencevideos)

- **Path**: `inputs.referenceVideos`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `5`
- **File size (each)**: `up to 100 MB`
- **Duration (each)**: `1s to 15s`
- **Duration (total)**: `up to 15s`
- **Size (each)**: `240 px or more`
- **Aspect ratio (each)**: `1:8 to 8:1`

List of reference videos (UUID or URL).

### [referenceAudios](https://runware.ai/docs/models/alibaba-wan3-0#request-inputs-referenceaudios)

- **Path**: `inputs.referenceAudios`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `5`
- **File size (each)**: `up to 15 MB`
- **Duration (each)**: `1s to 15s`
- **Duration (total)**: `up to 15s`

List of reference audios (UUID or URL).

### [documents](https://runware.ai/docs/models/alibaba-wan3-0#request-inputs-documents)

- **Path**: `inputs.documents`
- **Type**: `array of strings`
- **Min items**: `items: 1`
- **File size (each)**: `up to 100 MB`
- **Formats**: `docx, doc, xlsx, xls, pptx, ppt, key, pages, numbers, md`
- **Pages (each)**: `up to 50`

Array of document inputs (UUID, URL, or Base64).

### [urls](https://runware.ai/docs/models/alibaba-wan3-0#request-inputs-urls)

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

Array of public webpage URLs.

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/alibaba-wan3-0#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `alibaba:wan@3.0`

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/alibaba-wan3-0#request-positiveprompt)

- **Type**: `string`
- **Required**: conditional — Required when inputs.frameImages and inputs.referenceImages and inputs.referenceVideos and inputs.referenceAudios are not provided
- **Min**: `2`
- **Max**: `20000`

Text prompt describing elements to include in the generated output. Reference assets in array order as `Image 1`, `Video 1`, and `Audio 1`, numbered from one.

**Learn more** (1 resource):

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

### [width](https://runware.ai/docs/models/alibaba-wan3-0#request-width)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.frameImages and inputs.referenceImages and inputs.referenceVideos 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/alibaba-wan3-0#request-height)

- **Type**: `integer`
- **Required**: conditional — Required when inputs.frameImages and inputs.referenceImages and inputs.referenceVideos 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)

### [resolution](https://runware.ai/docs/models/alibaba-wan3-0#request-resolution)

- **Type**: `string`
- **Default**: `1080p`

Resolution preset for the output. When used with input media, automatically matches the aspect ratio from the input.

**Allowed values**: `480p` `720p` `1080p`

### [duration](https://runware.ai/docs/models/alibaba-wan3-0#request-duration)

- **Type**: `integer | string`
- **Min**: `2`
- **Max**: `30`
- **Default**: `auto`

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).

**Allowed values**: `auto`

### [seed](https://runware.ai/docs/models/alibaba-wan3-0#request-seed)

- **Type**: `integer`
- **Min**: `0`
- **Max**: `2147483647`

Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range.

**Settings**

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

### [audio](https://runware.ai/docs/models/alibaba-wan3-0#request-settings-audio)

- **Path**: `settings.audio`
- **Type**: `boolean`
- **Default**: `true`

Generate native audio aligned with visual content.

### [promptExtend](https://runware.ai/docs/models/alibaba-wan3-0#request-settings-promptextend)

- **Path**: `settings.promptExtend`
- **Type**: `boolean`
- **Default**: `true`

Enable LLM-based prompt rewriting to expand and clarify inputs. Affects reproducibility.

## Response Parameters

### [taskType](https://runware.ai/docs/models/alibaba-wan3-0#response-tasktype)

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

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0#response-videouuid)

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

UUID of the output video.

### [videoURL](https://runware.ai/docs/models/alibaba-wan3-0#response-videourl)

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

URL of the output video.

### [videoBase64Data](https://runware.ai/docs/models/alibaba-wan3-0#response-videobase64data)

- **Type**: `string`

Base64-encoded video data.

### [videoDataURI](https://runware.ai/docs/models/alibaba-wan3-0#response-videodatauri)

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

Data URI of the output video.

### [seed](https://runware.ai/docs/models/alibaba-wan3-0#response-seed)

- **Type**: `integer`

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

### [NSFWContent](https://runware.ai/docs/models/alibaba-wan3-0#response-nsfwcontent)

- **Type**: `boolean`

Flag indicating if NSFW content was detected.

### [cost](https://runware.ai/docs/models/alibaba-wan3-0#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/alibaba-wan3-0/42e55cf4-ecf1-4573-a28e-9941e4db302d.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "dd494c1a-8ec9-4fe7-83bd-97bd8056eaeb",
  "model": "alibaba:wan@3.0",
  "positivePrompt": "Create a polished 6-second landscape social media video ad for a mobile knife-sharpening service, set in a real restaurant prep kitchen after closing. Hyperreal tactile food-commercial cinematography with dark stainless steel, warm tungsten highlights, cool silver blades, deep shadows, and small flashes of ember orange. 0.0–1.4 seconds: extreme macro side view of a visibly dull chef’s knife crushing the skin of a ripe red tomato instead of cutting it; the cook’s hand is steady, clean, and professionally positioned. Camera makes a slow, tense push toward the buckling tomato skin. Sound: quiet kitchen room tone, refrigerator hum, rubber cutting board friction, and a soft wet squash. 1.4–3.8 seconds: sharp match cut to the same knife meeting a compact professional belt sharpener on a stainless workbench. Track left alongside the blade as the technician draws it smoothly across the moving belt; one controlled fan of bright sparks sweeps backward in slow motion. Sound synchronized precisely: electric motor rises, clean metallic grinding hiss, brief spark crackle. 3.8–6.0 seconds: match cut back to the prep board. The restored knife glides through a fresh tomato in one effortless stroke, producing a translucent, paper-thin slice that falls forward. Camera follows the blade with a fast low lateral slide, then settles on the pristine cut surface and gleaming edge. Sound: crisp blade whisper, delicate tap on the board, motor sound resolves into silence. Premium, practical, satisfying ASMR mood; confident commercial pacing; physically accurate hands, knife geometry, sparks, and food motion. No dialogue, captions, logos, labels, interface elements, or on-screen text.",
  "width": 1280,
  "height": 720,
  "duration": 6,
  "settings": {
    "audio": true
  }
}
```

---

### first-frame (first-frame)

[Watch video](https://assets.runware.ai/examples/alibaba-wan3-0/6de8e8d7-149c-429a-b8eb-ebe36e597928.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "170a0ba3-28e1-42ef-b66e-dccdb4dbf3a9",
  "model": "alibaba:wan@3.0",
  "positivePrompt": "Continue naturally from Image 1 as a single photorealistic documentary b-roll shot. The technician steadily lifts the intact pale mycelium packaging insert out of the stainless-steel mold, raises it several inches, then gently turns it toward the light to inspect its fibrous surface. A few tiny dry substrate crumbs fall back onto the workbench. Preserve the technician, gloves, mold, workshop layout, lighting and product geometry from the supplied first frame. Use a restrained, smooth lateral camera slide with subtle natural handheld presence, shallow but practical depth of field, realistic hand mechanics, accurate material weight and understated factory motion in the distant background. Muted cream, sage and rust palette; soft high-window daylight; tactile fibrous detail; premium editorial footage for a circular-manufacturing documentary. One continuous shot, no cuts, no slow motion, no dramatic reveal, no logos, no captions, no on-screen text.",
  "resolution": "720p",
  "duration": 5,
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/7ff25248-4c08-4148-9004-d861c4e292fc.jpg",
        "frame": "first"
      }
    ]
  }
}
```

---

### first-last-frame (first-last-frame)

[Watch video](https://assets.runware.ai/examples/alibaba-wan3-0/e74bc47b-735d-43bc-954c-4dc3441d213d.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "e811fa9d-c408-4671-ab61-2a7e402290eb",
  "model": "alibaba:wan@3.0",
  "positivePrompt": "Create one continuous eight-second cinematic shot transitioning coherently from Image 1 as the first frame to Image 2 as the last frame. Preserve the exact five falconers, their left-to-right positions, faces, proportions, navy workwear, service vehicle, runway background, camera height, focal length and cool-blue dawn lighting throughout. Begin as a composed airport wildlife-control crew portrait with the central woman holding up an empty falconry glove. After a quiet beat, the handlers glance naturally toward frame right as an adult Harris's hawk enters from the right, crosses the foreground with realistic wingbeats, slows, extends its talons and lands gently on the central handler's glove. Her arm responds subtly to the bird's weight. The hawk folds its wings while the team settles into the proud final group portrait shown in Image 2. Use a nearly locked camera with only a very subtle cinematic push-in. Keep motion physically plausible and documentary-realistic: accurate bird anatomy, stable human faces, natural blinking, slight jacket movement from the wing wash and faint mist drifting near the ground. No cuts, no dissolves, no teleportation, no extra people, no additional birds, no wardrobe changes, no logos and no on-screen text. Sound should feel authentic and restrained: distant runway ambience, light wind, soft wingbeats, leather glove movement and a faint hawk call.",
  "resolution": "1080p",
  "duration": 8,
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/c92d0697-c6f7-434b-b495-080ae9fb6c0f.jpg",
        "frame": "first"
      },
      {
        "image": "https://assets.runware.ai/assets/inputs/d633d15e-c95f-442d-bf4a-74f3fb591fc5.jpg",
        "frame": "last"
      }
    ]
  }
}
```

---

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

[Watch video](https://assets.runware.ai/examples/alibaba-wan3-0/8982a417-1cba-413f-9b06-777387c9abe3.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "6f8813b0-1051-4f73-bbff-544f219ad2fd",
  "model": "alibaba:wan@3.0",
  "positivePrompt": "Create a finished 10-second cinematic portrait title teaser for an investigative documentary about the hidden pneumatic tube network transporting urgent blood samples through a hospital at night. Image 1 defines the recurring subject: preserve the technician's identity, navy coveralls, clear safety glasses, scarlet specimen carrier, and the pale green basement engineering environment. Video 1 guides the physical motion and cinematography: reproduce the carrier's credible pneumatic acceleration, banking movement, strong tracking parallax, passage through junction hardware, and decisive arrival at a receiving station. Audio 1 guides the complete edit rhythm and sound design: synchronize cuts to its relay clicks and pressure pulses, build tension with its sub-bass rise, land the carrier on its sharp impact, and let the final air release accompany the title reveal.\n\nOpen in near darkness on an extreme macro view of the technician's eye reflected in polished tube hardware. Cut on the first relay click to her gloved hand loading the scarlet carrier. Follow it upward through a transparent tube in a fast vertical tracking shot designed for 9:16 framing; conduits, basement lights, and concrete floors streak past with controlled motion blur. Transition into a cutaway-like view through successive hospital levels: mechanical room, dim service corridor, then a sterile laboratory receiving bay, all photorealistic and architecturally coherent. Brief insert shots show pressure gauges twitching, routing gates snapping into place, and the technician looking upward as the capsule travels beyond sight. Use cold surgical cyan and institutional green, contrasted by the saturated scarlet carrier and occasional amber utility lamps. The mood is tense, intelligent, procedural, and quietly urgent, treating ordinary infrastructure like the circulatory system of the building.\n\nAt the carrier impact, cut to black for a fraction of a second. On the final air release, reveal the documentary title BETWEEN FLOORS in large condensed off-white uppercase lettering, centered and sharply legible over a dark close-up of branching pneumatic tubes. Beneath it, add only the small subtitle A DOCUMENTARY. No logos, credits, interface elements, watermarks, or other text. Premium documentary cinematography, realistic materials, restrained film grain, deep blacks, precise match cuts, coherent spatial continuity, polished broadcast-ready finish.",
  "width": 1080,
  "height": 1920,
  "duration": 10,
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/cafdfa1b-ffdd-450b-94fc-5a0758687665.jpg"
    ],
    "referenceVideos": [
      "https://assets.runware.ai/assets/inputs/cd97f896-a7b6-45d0-96d7-67cbddebe02d.mp4"
    ],
    "referenceAudios": [
      "https://assets.runware.ai/assets/inputs/138fe04a-cd34-47d2-8e1a-cfa481d7da74.mp3"
    ]
  }
}
```

---

### url-grounded (url-grounded)

[Watch video](https://assets.runware.ai/examples/alibaba-wan3-0/15dc7ee3-ee92-47fe-832b-d7841e889263.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "b96fbb6b-0254-47ee-b1f4-b3ab748b2610",
  "model": "alibaba:wan@3.0",
  "positivePrompt": "Create a polished 7-second square product colorway reveal using the supplied public product webpage as the authoritative reference for the Petzl METEOR helmet. Accurately preserve the helmet’s recognizable low-profile shell geometry, vent layout, brim, straps, adjustment system, surface materials, proportions, logos, and small functional details shown on the page. Change only the color treatment: recolor the main shell to a refined matte pale lichen green, with a deep aubergine lower rim and subtle aubergine adjustment details; keep straps charcoal gray and retain authentic existing markings without adding slogans or invented branding.\n\nBuild one coherent studio sequence rather than showing the webpage or any browser interface. Open on an extreme macro view of soft light sliding across the lichen shell and one sculpted ventilation opening. The camera pulls back as the helmet rotates smoothly on a dark plum stone pedestal, revealing the aubergine rim and accurate side profile. Finish with a clean three-quarter hero view as a narrow beam of cool alpine light crosses the helmet and the rotation settles. Background: abstract slate quarry forms softened by shallow depth of field, premium outdoor-equipment art direction, controlled mineral dust in the air, restrained motion, tactile matte surfaces, crisp product edges, realistic contact shadow. Palette limited to pale lichen, aubergine, charcoal, cool slate, and a trace of silver-white light. No people, no extra equipment, no packaging, no captions, no promotional text, no interface, no changes to product construction, and no morphing between shots.",
  "width": 960,
  "height": 960,
  "duration": 7,
  "inputs": {
    "urls": [
      "https://www.petzl.com/US/en/Sport/Helmets/METEOR"
    ]
  }
}
```