---
title: MiniMax H3 Max | Runware Docs
url: https://runware.ai/docs/models/minimax-h3-max
description: High-throughput video generation with stronger prompt adherence, native synced audio, and first-to-last frame control
---
# MiniMax H3 Max

![MiniMax H3 Max](https://assets.runware.ai/covers/minimax-h3-max.jpg)

MiniMax H3 Max is a performance-tuned variant of MiniMax H3 built for faster video generation while preserving strong prompt following and polished audiovisual results. It supports text-to-video and image-to-video workflows, including optional first-and-last-frame guidance for more controlled motion between key images, and is well suited to rapid ideation, high-volume content production, and interactive creative workflows that need lower latency without giving up visual quality.

- **Status**: live
- **ID**: `minimax:h3@max`
- Try in Playground
- **Creator**: MiniMax
- **Release Date**: August 26, 2026
- **Zero Data Retention**: Yes
- **Capabilities**: Text to Video, Image to Video, Video to Video, Audio to Video, Checkpoint, Audio

## Pricing

$0.05 from, per second · ~24s per request

**Rates** (per second)

- 480p: `$0.05`
- 768p: `$0.08`

**4 example generations**

- [first-frame](https://runware.ai/docs/models/minimax-h3-max/examples#example-vineyard-falconry-social-video-ad) · 768p · 5s: `$0.40` (~23s)
- [first-last-frame](https://runware.ai/docs/models/minimax-h3-max/examples#example-accordion-restoration-reel-ad) · 480p · 15s: `$0.75` (~25s)
- [text-to-video](https://runware.ai/docs/models/minimax-h3-max/examples#example-mycelium-biomaterial-lab-b-roll) · 1344×768 · 10s: `$0.80` (~24s)
- [text-to-video](https://runware.ai/docs/models/minimax-h3-max/examples#example-museum-seismic-plinth-explainer) · 1344×768 · 10s: `$0.80` (~23s)

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

## Compatibility & Validation

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

---

When `inputs.frameImages` is provided, `width/height` cannot be used.

---

When `inputs.referenceAudios` is provided, `inputs.referenceImages` or `inputs.referenceVideos` is required.

---

When `resolution` is provided, `inputs.frameImages` is required.

---

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

---

The following dimension combinations are supported:

| Configuration | Dimensions |
| --- | --- |
| `480p (~16:9)` | `832x480` |
| `480p (4:3)` | `640x480` |
| `480p (1:1)` | `480x480` |
| `480p (3:4)` | `480x640` |
| `480p (~9:16)` | `480x832` |
| `480p (~21:9)` | `960x416` |
| `768p (~16:9)` | `1344x768` |
| `768p (4:3)` | `1024x768` |
| `768p (1:1)` | `768x768` |
| `768p (3:4)` | `768x1024` |
| `768p (~9:16)` | `768x1344` |
| `768p (~21:9)` | `1536x672` |

## Request Parameters

**API Options**

Platform-level options for task execution and delivery.

### [taskType](https://runware.ai/docs/models/minimax-h3-max#request-tasktype)

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

Identifier for the type of task being performed.

### [taskUUID](https://runware.ai/docs/models/minimax-h3-max#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/minimax-h3-max#request-outputtype)

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

Video output type.

**Allowed values**: `URL`

### [outputFormat](https://runware.ai/docs/models/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#request-safety)

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

Content safety checking configuration for video generation.

#### [checkContent](https://runware.ai/docs/models/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#request-includecost)

- **Type**: `boolean`

Include task cost in the response.

### [numberResults](https://runware.ai/docs/models/minimax-h3-max#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/minimax-h3-max#request-inputs-referenceimages)

- **Path**: `inputs.referenceImages`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `9`
- **Size (each)**: `256 px to 5760 px`
- **Aspect ratio (each)**: `1:2.5 to 2.5:1`
- **File size (each)**: `up to 30 MB`

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

### [frameImages](https://runware.ai/docs/models/minimax-h3-max#request-inputs-frameimages)

- **Path**: `inputs.frameImages`
- **Type**: `array of strings or objects`
- **Min items**: `1`
- **Max items**: `2`
- **Size (each)**: `256 px to 5760 px`
- **Aspect ratio (each)**: `1:2.5 to 2.5:1`
- **File size (each)**: `up to 30 MB`

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/minimax-h3-max#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/minimax-h3-max#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/minimax-h3-max#request-inputs-referencevideos)

- **Path**: `inputs.referenceVideos`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `3`
- **Size (each)**: `256 px to 5760 px`
- **Aspect ratio (each)**: `1:2.5 to 2.5:1`
- **Frame rate (each)**: `23.98 fps to 60 fps`
- **Duration (each)**: `2s to 15s`
- **Duration (total)**: `up to 15s`
- **File size (each)**: `up to 50 MB`

List of reference videos (UUID or URL).

### [referenceAudios](https://runware.ai/docs/models/minimax-h3-max#request-inputs-referenceaudios)

- **Path**: `inputs.referenceAudios`
- **Type**: `array of strings`
- **Min items**: `1`
- **Max items**: `3`
- **Duration (each)**: `2s to 15s`
- **Duration (total)**: `up to 15s`
- **File size (each)**: `up to 15 MB`

List of reference audios (UUID or URL).

**Core Parameters**

Primary parameters that define the task output.

### [model](https://runware.ai/docs/models/minimax-h3-max#request-model)

- **Type**: `string`
- **Required**: true
- **Value**: `minimax:h3@max`

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/minimax-h3-max#request-positiveprompt)

- **Type**: `string`
- **Required**: true
- **Min**: `2`
- **Max**: `7000`

Text prompt describing elements to include in the generated output.

**Learn more** (1 resource):

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

### [width](https://runware.ai/docs/models/minimax-h3-max#request-width)

- **Type**: `integer`
- **Default**: `1344`
- **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/minimax-h3-max#request-height)

- **Type**: `integer`
- **Default**: `768`
- **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/minimax-h3-max#request-resolution)

- **Type**: `string`

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

**Allowed values**: `480p` `768p`

### [duration](https://runware.ai/docs/models/minimax-h3-max#request-duration)

- **Type**: `integer`
- **Min**: `5`
- **Max**: `15`
- **Default**: `5`

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).

### [seed](https://runware.ai/docs/models/minimax-h3-max#request-seed)

- **Type**: `integer`
- **Min**: `0`
- **Max**: `4294967295`

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.

### [promptExpansion](https://runware.ai/docs/models/minimax-h3-max#request-settings-promptexpansion)

- **Path**: `settings.promptExpansion`
- **Type**: `string`
- **Default**: `balanced`

Level of automatic prompt rewriting applied before generation. Affects reproducibility.

**Allowed values**:

- `disabled` No prompt rewriting.
- `balanced` Balanced rewriting.
- `quality` Quality-first rewriting, at the cost of longer generation time.

## Response Parameters

### [taskType](https://runware.ai/docs/models/minimax-h3-max#response-tasktype)

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

Identifier for the type of task this response belongs to.

### [taskUUID](https://runware.ai/docs/models/minimax-h3-max#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/minimax-h3-max#response-videouuid)

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

UUID of the output video.

### [videoURL](https://runware.ai/docs/models/minimax-h3-max#response-videourl)

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

URL of the output video.

### [videoBase64Data](https://runware.ai/docs/models/minimax-h3-max#response-videobase64data)

- **Type**: `string`

Base64-encoded video data.

### [videoDataURI](https://runware.ai/docs/models/minimax-h3-max#response-videodatauri)

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

Data URI of the output video.

### [seed](https://runware.ai/docs/models/minimax-h3-max#response-seed)

- **Type**: `integer`

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

### [NSFWContent](https://runware.ai/docs/models/minimax-h3-max#response-nsfwcontent)

- **Type**: `boolean`

Flag indicating if NSFW content was detected.

### [cost](https://runware.ai/docs/models/minimax-h3-max#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/minimax-h3-max/5818e405-0da7-41b1-9b2c-b26100765bd8.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "0e0bdf36-080b-40ff-8d70-18249b6292d9",
  "model": "minimax:h3@max",
  "positivePrompt": "A polished 10-second documentary b-roll shot inside a real mycelium biomaterials laboratory. A materials technician wearing a pale sage lab coat, hair covering, nitrile gloves, and clear safety glasses carefully lifts a flexible cream-colored sheet of grown mycelium from a shallow stainless-steel cultivation tray. The fibrous material resembles unfinished plant-based leather, with subtle organic texture and irregular natural edges. Surrounding environment: clean stainless workbench, stacked incubation trays, glass partitions, soft vapor from a nearby humidification system, and defocused laboratory equipment in the background. Warm ivory, muted sage, and brushed-steel palette; soft diffused overhead lighting with gentle morning light entering from one side. Begin with a stable waist-level medium shot showing the technician and tray. During the first three seconds, the technician loosens one edge with a small sterile spatula. From seconds three to seven, they slowly raise the supple sheet with both hands as it bends naturally under gravity. The camera performs a smooth, subtle lateral dolly and gradual push-in, keeping the hands and material in sharp focus while the background falls into shallow depth of field. During the final three seconds, the technician holds the sheet toward a soft inspection light, revealing its translucent fibers and fine surface texture; finish on a clean close-up with gentle vapor drifting through the backlight. One continuous shot, physically realistic hand movement and material behavior, restrained professional stock-footage aesthetic, no cuts, no dramatic visual effects, no visible logos, labels, screens, captions, or text.",
  "width": 1344,
  "height": 768,
  "duration": 10
}
```

---

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

[Watch video](https://assets.runware.ai/examples/minimax-h3-max/5fe73e6e-bd9e-4b67-9916-8c9540126e20.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "806d3864-5854-4cd9-8f24-eda82fec8ebf",
  "model": "minimax:h3@max",
  "positivePrompt": "Create a polished 10-second photorealistic product explainer for a museum engineering supplier. The product is an unbranded seismic isolation plinth: a low, precision-machined circular pedestal with a pale limestone top, dark graphite base, and a thin vermilion accent ring. A delicate ancient celadon vase stands centered on top. The demonstration takes place in an immaculate museum conservation studio with cool gray walls, archival shelving softly out of focus, and warm directional daylight.\n\n0–2 seconds: begin with a low three-quarter hero view and a slow, steady dolly toward the vase and plinth. The pedestal sits on a compact black earthquake simulation table. Crisp reflections and shallow depth of field emphasize the engineered materials.\n\n2–6 seconds: the simulation table suddenly oscillates left and right with forceful, clearly visible earthquake motion. The black table and lower section of the plinth shift rapidly while the isolated limestone top glides in controlled counter-motion. The tall vase remains upright, calm, and nearly motionless, making the protection immediately understandable. A small shallow dish of water beside the vase stays level with only a faint ripple. Motion must feel mechanically credible, not magical.\n\n6–8 seconds: the shaking decelerates and stops. The plinth smoothly recenters itself while the vase remains perfectly safe. The camera begins a subtle clockwise arc around the product.\n\n8–10 seconds: during the arc, the dark outer casing becomes a refined semi-transparent technical cutaway, revealing crossed steel guide rails, circular bearings, and compact spring dampers returning to center beneath the stable top plate. End on a clean three-quarter close-up of the exposed mechanism and protected vase. Elegant museum-grade industrial design, celadon green, limestone cream, graphite, and vermilion palette; precise premium engineering-film aesthetic. Natural synchronized sound: low shake-table rumble, controlled mechanical sliding, then a soft settling click. No people, captions, logos, interface graphics, diagrams, or floating text.",
  "width": 1344,
  "height": 768,
  "duration": 10
}
```

---

### first-frame (first-frame)

[Watch video](https://assets.runware.ai/examples/minimax-h3-max/6d60c2f0-c958-4231-874d-4dfe124e8e06.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "a4483454-dc7b-4080-a356-407de7e883e3",
  "model": "minimax:h3@max",
  "positivePrompt": "Continue naturally from the supplied first frame as one polished vertical social media ad shot. The falconer makes a subtle forward gesture with her raised gauntlet and the Harris's hawk pushes off powerfully, opening its wings with realistic weight and feather movement. Loose grape leaves stir from the downbeat. The camera tilts and glides upward to follow the hawk flying low along the vineyard row, keeping it sharp as the handler and vines fall into gentle background blur. A small flock of pest birds lifts safely from the distant vines and disperses ahead of the approaching hawk, communicating humane crop protection without aggression or contact. Preserve the cool slate-blue dawn, copper rim light, thin ground mist, burgundy jacket and refined agricultural-documentary art direction. Smooth physically plausible motion, restrained slow-motion feeling, subtle handheld steadiness, realistic wing beats, cinematic depth of field, premium editorial finish. Keep all people, birds and vineyard structures natural and consistent. No cuts, captions, logos, interface elements or added text.",
  "resolution": "768p",
  "duration": 5,
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/cd80e1f1-e7c7-4786-9fc1-557ebcc635f5.jpg",
        "frame": "first"
      }
    ]
  }
}
```

---

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

[Watch video](https://assets.runware.ai/examples/minimax-h3-max/c506d025-8294-4986-a996-a11d79343364.mp4)

```json
{
  "taskType": "videoInference",
  "taskUUID": "a2d362d6-1196-4303-b4f0-3b6683bf4be1",
  "model": "minimax:h3@max",
  "positivePrompt": "Create a polished 15-second vertical short-form social ad for a specialist accordion restoration atelier. Begin precisely from the supplied first frame and finish precisely on the supplied last frame, preserving the same vintage accordion, locked three-quarter camera angle, walnut workbench, warm task lamp and teal workshop shadows throughout. The visual language is tactile, rhythmic and premium rather than whimsical.\n\n0-3 seconds: a narrow beam from the task lamp sweeps across the dusty, damaged accordion. Fine dust lifts into the light as the camera makes an extremely subtle push forward. The sagging bellows exhale with a soft, tired wheeze. A craftsperson's steady hands in an indigo work apron enter frame and gently test one crooked key.\n\n3-11 seconds: compress the restoration process into a sequence of elegant, physically believable match-cut motions while maintaining spatial continuity. Hands release tiny screws into a brass tray; lift the casing; remove and inspect tarnished reeds; brush dust from the interior; straighten the key mechanism with fine pliers; apply wax; patch and press the bellows corners; stitch the torn leather strap; polish the oxblood-red marbled shell with a soft cloth. Use close, satisfying hand movements, brief rack focuses, restrained speed ramps and seamless object-led transitions. Tools move only when handled. Materials react realistically: dust disperses, metal catches highlights, wax softens, leather flexes and polish gradually restores a deep gloss. Do not introduce sparks, magical particles or impossible morphing.\n\n11-15 seconds: return smoothly to the original wider composition as the finished accordion settles into the exact final-frame position. The craftsperson's hands lift and expand the renewed bellows, then press a clean chord. End on the supplied restored image with the polished casing glowing under the tungsten lamp and the bellows gently breathing. Sound design evolves from dry clicks, cloth friction and the initial weak wheeze into crisp key action and one warm, resonant accordion chord. Rich oxblood, aged ivory, brass, walnut and deep teal palette; cinematic workshop realism; refined craft documentary art direction; stable anatomy and consistent object geometry. No captions, logos, readable lettering, borders, interface elements or watermarks.",
  "resolution": "480p",
  "duration": 15,
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/6ab315f7-edba-4a00-9c10-97c876e5d075.jpg",
        "frame": "first"
      },
      {
        "image": "https://assets.runware.ai/assets/inputs/4a6a0a43-c347-4dee-b052-dd8a691e466d.jpg",
        "frame": "last"
      }
    ]
  }
}
```