### Rainy Museum Audio Guide

Source: https://runware.ai/docs/models/alibaba-qwen3-tts-1-7b-base.md

Generates an audio guide for a museum exhibit, describing an astrolabe and its historical context. This example highlights descriptive text and specific audio parameters.

```json
{
  "taskType": "audioInference",
  "taskUUID": "58703dfd-8d1b-46f2-8efd-8ef2a9f23fdd",
  "model": "alibaba:qwen@3-tts-1.7b-base",
  "speech": {
    "text": "Good evening, and welcome to Gallery Seven. Outside, the rain taps softly against the tall glass panes, but in here the room stays hushed, warm, and full of suspended attention. Before you is a weathered bronze astrolabe recovered from a river crossing centuries ago. Notice the worn edge near the lower ring; that smoothness came not from time alone, but from repeated handling by navigators who trusted it during long journeys. As you move closer, look for the tiny engraved star marks arranged in uneven arcs. They are easy to miss at first, yet they reveal how practical knowledge and artistry once lived side by side. When you are ready, continue to the next display, where a series of field journals records the same route in three different seasons.",
    "voice": "clone",
    "language": "English",
    "speed": 0.93
  },
  "settings": {
    "maxNewTokens": 2048,
    "transcript": "Welcome. Please follow me to the next gallery for a closer look.",
    "xVectorOnly": false
  },
  "inputs": {
    "audio": "https://assets.runware.ai/assets/inputs/6e19ca15-d4dc-4a2e-ba62-6d39e7d2e8b5.mp3"
  }
}
```

```json
{
  "taskType": "audioInference",
  "taskUUID": "58703dfd-8d1b-46f2-8efd-8ef2a9f23fdd",
  "audioUUID": "35e2a7b7-1825-4cb4-9d75-4ed8281d6cd4",
  "audioURL": "https://am.runware.ai/audio/os/a21d05/ws/5/ai/35e2a7b7-1825-4cb4-9d75-4ed8281d6cd4.wav",
  "cost": 0.0008
}
```

--------------------------------

### Example Request and Response

Source: https://runware.ai/docs/models/klingai-video-o3-pro.md

This example demonstrates a typical request to generate a video using the KlingAI Video O3 Pro model and its corresponding successful response.

```APIDOC
## Examples

### Windblown Tundra Signal Camp (Text to Video)

**Request**: 
```json
{
  "taskType": "videoInference",
  "taskUUID": "f92fd87a-7215-449a-8281-7be52ed8f447",
  "model": "klingai:kling-video@o3-pro",
  "positivePrompt": "A cinematic arctic field camp at blue-hour twilight on a vast frozen plain, a lone polar researcher in a crimson expedition parka stands beside a tripod radio mast and a small tracked snow vehicle, distant ice ridges and drifting ground snow, nylon flags snapping hard in the wind, frost vapor from breathing, the researcher adjusts a headset and aims a directional antenna toward the horizon, then a faint green aurora ribbon slowly appears overhead while the machine idles and indicator lights blink, ultra-detailed realistic textures, natural body movement, believable wind interaction, dramatic wide shot with a slow lateral camera glide, high motion realism, crisp atmospheric depth, immersive native environmental audio with wind, engine hum, fabric flapping, radio static",
  "width": 1920,
  "height": 1080,
  "duration": 8,
  "providerSettings": {
    "klingai": {
      "sound": true
    }
  }
}
```

**Response**: 
```json
{
  "taskType": "videoInference",
  "taskUUID": "f92fd87a-7215-449a-8281-7be52ed8f447",
  "videoUUID": "0d284b5f-9714-4731-a03f-5dd1afeeb4f6",
  "videoURL": "https://vm.runware.ai/video/os/a19d05/ws/5/vi/0d284b5f-9714-4731-a03f-5dd1afeeb4f6.mp4",
  "cost": 1.12
}
```
```

--------------------------------

### ControlNet Preprocess Canny Response Example 4

Source: https://runware.ai/docs/models/controlnet-preprocess-canny/examples

The final example response from the ControlNet Preprocess Canny model, providing the guide image URL and UUID for the sculpture input.

```json
{
  "taskType": "imageControlNetPreProcess",
  "taskUUID": "963bfadc-61cf-4e50-a90f-2ae9b4ab1cda",
  "guideImageURL": "https://im.runware.ai/image/os/a06dlim3/ws/3/ii/28acbf79-ad81-437b-86e3-974fee8158e4.jpg",
  "cost": 0.0006,
  "guideImageUUID": "28acbf79-ad81-437b-86e3-974fee8158e4",
  "inputImageUUID": "8e42795e-f465-4719-9931-ebc4e7b1efc8"
}
```

--------------------------------

### Stormy Lighthouse Rescue Approach (Image to Video) Example

Source: https://runware.ai/docs/models/ovi.md

Example request and response for generating a video from an image with a detailed prompt.

```APIDOC
## Stormy Lighthouse Rescue Approach (Image to Video)

### Request Example
```json
{
  "taskType": "videoInference",
  "taskUUID": "6537bd8a-b38e-4579-9b6f-7f48cd2f4f38",
  "model": "runware:190@1",
  "positivePrompt": "Animate this storm-battered lighthouse scene into a high-tension cinematic rescue sequence. The helicopter advances through heavy rain and crosswinds, rotor blades slicing mist into spiraling sheets. The lighthouse beacon sweeps across the sea and briefly catches the aircraft fuselage. Waves hammer the rocks in irregular bursts, white spray rising with each impact. Lightning flashes intermittently, revealing texture in the clouds and ocean surface. Camera feel: slow forward drift with subtle handheld turbulence, as if filmed from another aircraft nearby. Motion should feel physically grounded and coherent: rain driven sideways by wind, foam dragged back into the surf, helicopter navigation lights pulsing through the storm, searchlight glancing across wet stone. Audio should emerge naturally from the scene with synchronized thunder cracks, rotor thrum, wind gusts, rain striking metal, distant warning siren from the lighthouse, and booming surf. Moody, realistic, suspenseful, immersive.",
  "fps": 24,
  "seed": 79445,
  "steps": 40,
  "inputs": {
    "image": "https://assets.runware.ai/assets/inputs/597fde48-a2bd-4a5b-a57d-2e8673b348e1.jpg"
  }
}
```

### Response Example
```json
{
  "taskType": "videoInference",
  "taskUUID": "6537bd8a-b38e-4579-9b6f-7f48cd2f4f38",
  "videoUUID": "885007e4-58d3-42e8-9bca-5fa150e75eb7",
  "videoURL": "https://vm.runware.ai/video/os/a05d22/ws/5/vi/885007e4-58d3-42e8-9bca-5fa150e75eb7.mp4",
  "seed": 79445,
  "cost": 0.1472
}
```
```

--------------------------------

### Text to Image Request Example

Source: https://runware.ai/docs/models/ideogram-3-0.md

Example of a request to generate an image with a specific prompt, dimensions, and model. This includes detailed positive and negative prompts to guide the image generation and ensure typography accuracy.

```json
{
  "taskType": "imageInference",
  "taskUUID": "3a300027-be07-4fd4-a95b-4f1b8dd572ef",
  "model": "ideogram:4@1",
  "positivePrompt": "A richly detailed underground subway jazz club entrance transformed into an elegant live-music venue, warm amber bulbs, tiled walls, polished brass handrails, velvet rope, hand-painted sandwich board, framed gig flyers, and a large illuminated marquee with crisp readable text: 'THE COPPER NOTE' on the main sign, smaller text below: 'LIVE TONIGHT' and 'TRIO AT 8:30'. A diverse crowd in stylish evening attire gathers on the platform, saxophone case near the stairs, subtle haze, glossy floor reflections, dramatic perspective down the tunnel, photorealistic textures, sophisticated atmosphere, impeccable typography, balanced composition, high detail.",
  "negativePrompt": "blurry text, misspelled words, duplicated people, extra limbs, warped hands, unreadable signage, low detail, oversaturated colors, empty scene, cartoonish rendering",
  "width": 1280,
  "height": 704
}
```

--------------------------------

### Image to Video Example

Source: https://runware.ai/docs/models/alibaba-wan2-7/examples

Generates a video from a reference image and a text prompt using the Wan2.7 model. This example shows how to include reference images and other parameters for image-to-video generation.

```APIDOC
## Image to Video

### Description
Generates a video from a reference image and a text prompt. This operation allows for animating a scene based on a provided image, maintaining character and composition consistency.

### Method
POST

### Endpoint
/video/inference

### Parameters
#### Request Body
- **taskType** (string) - Required - Must be "videoInference".
- **taskUUID** (string) - Required - Unique identifier for the task.
- **model** (string) - Required - The model to use, e.g., "alibaba:wan@2.7".
- **positivePrompt** (string) - Required - Detailed description of the desired video content, referencing the input image.
- **negativePrompt** (string) - Optional - Description of elements to exclude from the video.
- **width** (integer) - Required - The width of the output video in pixels.
- **height** (integer) - Required - The height of the output video in pixels.
- **duration** (integer) - Required - The duration of the video in seconds.
- **seed** (integer) - Optional - Seed for reproducible results.
- **settings** (object) - Optional - Additional settings.
  - **audio** (boolean) - Optional - Whether to generate audio for the video.
- **inputs** (object) - Required - Input data for the task.
  - **referenceImages** (array of strings) - Required - URLs of reference images to guide video generation.

### Request Example
```json
{
  "taskType": "videoInference",
  "taskUUID": "0d84f171-2e54-4c07-836e-05788795736d",
  "model": "alibaba:wan@2.7",
  "positivePrompt": "Using the reference image as the main visual anchor, animate a gentle departure scene: steam rolls across the platform, the balloon ropes tighten and flutter, the child in the cobalt coat steps forward and looks up, porters adjust luggage, the locomotive releases a soft burst of vapor, then the balloon lifts slowly into the air as the camera rises with it, revealing snowy pine ridges and a glowing dawn horizon. Preserve the character design, wardrobe, and overall composition from the reference. Realistic motion, cinematic depth, crisp cold atmosphere, delicate fabric movement, drifting snow powder, warm brass details against icy surroundings, elegant and emotional travel mood.",
  "negativePrompt": "low detail, distorted anatomy, extra limbs, duplicate people, flicker, jitter, warped face, text, watermark, oversaturated, chaotic motion, blurry subject",
  "width": 1280,
  "height": 720,
  "duration": 8,
  "seed": 9669,
  "settings": {
    "audio": true
  },
  "inputs": {
    "referenceImages": [
      "https://assets.runware.ai/assets/inputs/41851527-3df1-41c9-a39b-a0d358186d9f.jpg"
    ]
  }
}
```

### Response
#### Success Response (200)
- **taskType** (string) - The type of task performed.
- **taskUUID** (string) - The unique identifier for the task.
- **videoUUID** (string) - The unique identifier for the generated video.
- **videoURL** (string) - The URL to access the generated video.
- **seed** (integer) - The seed used for generation.
- **cost** (number) - The cost associated with the generation.

#### Response Example
```json
{
  "taskType": "videoInference",
  "taskUUID": "0d84f171-2e54-4c07-836e-05788795736d",
  "videoUUID": "9b2d1445-99ad-40ca-9adc-2dc6786df569",
  "videoURL": "https://vm.runware.ai/video/os/a07d11/ws/5/vi/9b2d1445-99ad-40ca-9adc-2dc6786df569.mp4",
  "seed": 9669,
  "cost": 0.8034
}
```
```

--------------------------------

### Image to Text Caption Example 1

Source: https://runware.ai/docs/models/alibaba-qwen2-5-vl-7b-instruct/examples

Generates a rich caption for an image, identifying the setting, human activity, background details, and mood. This example uses a specific prompt to guide the caption generation.

```json
{
  "taskType": "caption",
  "taskUUID": "eb705042-61a2-4922-8945-642ebf948d57",
  "model": "runware:152@2",
  "prompt": "Write a rich caption for this image in 2-3 sentences. Identify the primary setting, the main human activity, notable background details, and the overall mood. Keep the description grounded in visible evidence and avoid speculation.",
  "inputs": {
    "image": "https://assets.runware.ai/assets/inputs/ddf33190-1c7c-4b47-9207-d219fdd0491c.jpg"
  }
}
```

```json
{
  "taskType": "imageCaption",
  "taskUUID": "eb705042-61a2-4922-8945-642ebf948d57",
  "text": "The image captures a serene waterfront scene where three individuals are engaged in different activities. The central figure, wearing an apron, is meticulously arranging small cups on a table, suggesting he might be preparing tea or coffee. To his right, two men are seated at another table, engrossed in examining a map, possibly planning a journey or discussing directions. The backdrop features a misty harbor filled with docked fishing boats, adding a sense of calm and anticipation to the setting.",
  "cost": 0.0019
}
```

--------------------------------

### Example Request and Response

Source: https://runware.ai/docs/models/lightricks-ltx-2-pro.md

An example of a request to generate a video with the Lightricks LTX 2 Pro model and its corresponding response.

```APIDOC
## Examples

### Moonlit Bioluminescent Cliff Shrine (Image to Video)

**Request**:

```json
{
  "taskType": "videoInference",
  "taskUUID": "81c6d9e9-60fd-40ed-b6a6-fffbd534d476",
  "model": "lightricks:2@0",
  "positivePrompt": "Using the provided first frame as the opening shot, create a cinematic night sequence of a secluded shrine on a rugged sea cliff. The camera begins with a slow forward drift and slight parallax around the lantern-bearing traveler, while moonlight catches the shrine roof, wet stone, and wind-tossed grasses. Bioluminescent flowers pulse softly in clusters, fog rolls along the cliff edge, and waves crash far below with occasional spray. Thin clouds slide across the moon, casting shifting silver light and delicate shadows. Preserve the composition and subject identity from the frame image while adding realistic environmental motion, subtle cloth movement, gentle lantern flicker, and polished filmic pacing. Naturalistic textures, high dynamic range, precise lighting, immersive ambience, premium cinematic realism.",
  "width": 1920,
  "height": 1080,
  "duration": 8,
  "fps": 25,
  "providerSettings": {
    "lightricks": {
      "generateAudio": true
    }
  },
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/085e3f32-4bb6-445a-9e45-309f3d838ae6.jpg",
        "frame": "first"
      }
    ]
  }
}
```

**Response**:

```json
{
  "taskType": "videoInference",
  "taskUUID": "81c6d9e9-60fd-40ed-b6a6-fffbd534d476",
  "videoUUID": "32c216d2-8292-46ce-b520-c9a03aac5606",
  "videoURL": "https://vm.runware.ai/video/os/a15d18/ws/5/vi/32c216d2-8292-46ce-b520-c9a03aac5606.mp4",
  "cost": 0.48
}
```
```

--------------------------------

### Generate Video with Audio and Input Frame

Source: https://runware.ai/docs/models/lightricks-ltx-2-pro.md

This example demonstrates how to generate a video using the Lightricks LTX 2 Pro model. It includes specifying the task type, a unique task UUID, the model to use, a detailed positive prompt, video dimensions, duration, FPS, and provider-specific settings like enabling audio generation. It also shows how to provide an initial frame image for the video sequence.

```json
{
  "taskType": "videoInference",
  "taskUUID": "81c6d9e9-60fd-40ed-b6a6-fffbd534d476",
  "model": "lightricks:2@0",
  "positivePrompt": "Using the provided first frame as the opening shot, create a cinematic night sequence of a secluded shrine on a rugged sea cliff. The camera begins with a slow forward drift and slight parallax around the lantern-bearing traveler, while moonlight catches the shrine roof, wet stone, and wind-tossed grasses. Bioluminescent flowers pulse softly in clusters, fog rolls along the cliff edge, and waves crash far below with occasional spray. Thin clouds slide across the moon, casting shifting silver light and delicate shadows. Preserve the composition and subject identity from the frame image while adding realistic environmental motion, subtle cloth movement, gentle lantern flicker, and polished filmic pacing. Naturalistic textures, high dynamic range, precise lighting, immersive ambience, premium cinematic realism.",
  "width": 1920,
  "height": 1080,
  "duration": 8,
  "fps": 25,
  "providerSettings": {
    "lightricks": {
      "generateAudio": true
    }
  },
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/085e3f32-4bb6-445a-9e45-309f3d838ae6.jpg",
        "frame": "first"
      }
    ]
  }
}
```

```json
{
  "taskType": "videoInference",
  "taskUUID": "81c6d9e9-60fd-40ed-b6a6-fffbd534d476",
  "videoUUID": "32c216d2-8292-46ce-b520-c9a03aac5606",
  "videoURL": "https://vm.runware.ai/video/os/a15d18/ws/5/vi/32c216d2-8292-46ce-b520-c9a03aac5606.mp4",
  "cost": 0.48
}
```

--------------------------------

### Museum Planetarium Audio Guide Request

Source: https://runware.ai/docs/models/alibaba-qwen3-tts-1-7b-voicedesign.md

Example request for generating an audio guide with a specific voice profile. Includes task type, model, detailed positive prompt, speech content, and generation settings.

```json
{
  "taskType": "audioInference",
  "taskUUID": "7a687894-17db-4452-a7c4-87929a5f31a7",
  "model": "alibaba:qwen@3-tts-1.7b-voicedesign",
  "positivePrompt": "A composed, knowledgeable female voice in her forties with a clear international English accent, gentle resonance, measured pacing, and soft breath control. The tone is welcoming and intellectually curious, like a senior museum guide leading visitors beneath a domed star theater. Subtle awe on cosmic phrases, crisp diction on scientific terms, never theatrical, never robotic. Warm, polished, intimate public-speaking quality with natural pauses and graceful sentence endings.",
  "speech": {
    "text": "Welcome to the Hall of Ancient Skies. Above you, the ceiling recreates the winter constellations as they were charted over a thousand years ago. Notice how Orion stands slightly tilted compared with modern maps; this projection follows the perspective used by navigators who crossed open water with nothing but memory, geometry, and nerve. To your left is a bronze armillary sphere, its rings modeling the heavens as a set of interlocking paths. It is not only an instrument, but a theory made visible. As the lights dim, watch the red point near the horizon. That is Mars, rendered as it appeared during a famously bright opposition. For many cultures, such moments were more than astronomical events. They were signals, arguments, omens, and invitations to ask a daring question: if the sky changes, what else might change with it?",
    "voice": "design",
    "language": "English",
    "speed": 0.94
  },
  "settings": {
    "maxNewTokens": 2048
  }
}
```

--------------------------------

### Generating Content with Specific Instructions

Source: https://runware.ai/docs/models/google-gemini-3-5-flash/examples

Illustrates how to provide detailed instructions to the Gemini model for more controlled content generation. Use clear prompts for best results.

```python
import google.generativeai as genai

genai.configure(api_key="YOUR_API_KEY")

model = genai.GenerativeModel('gemini-3.5-flash')

response = model.generate_content(
    "Write a short poem about a robot learning to dream, in the style of Edgar Allan Poe."
)
print(response.text)
```

--------------------------------

### ControlNet Preprocess Scribble Response Example

Source: https://runware.ai/docs/models/controlnet-preprocess-scribble/examples

Example JSON response from the ControlNet Preprocess Scribble model. It includes the task type, task UUID, a URL for the generated guide image, cost, and image UUIDs.

```json
{
  "taskType": "imageControlNetPreProcess",
  "taskUUID": "f924796e-ee89-4909-9d66-45ab9c95193f",
  "guideImageURL": "https://im.runware.ai/image/os/a05d22/ws/3/ii/b3c083a6-0b41-4a36-b408-af4df064df31.jpg",
  "cost": 0.0006,
  "guideImageUUID": "b3c083a6-0b41-4a36-b408-af4df064df31",
  "inputImageUUID": "14184e58-25c6-47cc-bef8-c0b050505a31"
}
```

--------------------------------

### Generate Dutch Audio Guide

Source: https://runware.ai/docs/models/elevenlabs-multilingual-v2.md

Example of generating an audio guide in Dutch using the ElevenLabs model. Includes request parameters for text, language, and voice settings, and shows the expected response with audio URL and cost.

```json
{
  "taskType": "audioInference",
  "taskUUID": "35b9427c-7f64-45ef-94fa-325cadc9b243",
  "model": "elevenlabs:26@1",
  "seed": 2213,
  "speech": {
    "text": "Welkom in Utrecht. Voor je zie je de werfkelders langs de gracht, ooit gebruikt voor opslag en handel, en nu gevuld met kleine cafés, ateliers en muziek. Loop rustig verder over de kade. Aan je rechterhand hoor je misschien een fietsbel, en ergens in de verte slaat een kerkklok het uur. Deze stad beweegt in een vriendelijk tempo: levendig, maar nooit gehaast. Als je even stilstaat bij de brug, kijk dan naar het water. De weerspiegeling van de oude gevels maakt zelfs een gewone middag bijzonder. Over een paar minuten bereik je de Domtoren, het bekendste herkenningspunt van de stad.",
    "language": "nl"
  },
  "providerSettings": {
    "elevenlabs": {
      "textToSpeech": {
        "voiceSettings": {
          "stability": 0.42,
          "similarityBoost": 0.78,
          "style": 0.61,
          "useSpeakerBoost": true
        },
        "applyTextNormalization": "auto"
      }
    }
  }
}
```

```json
{
  "taskType": "audioInference",
  "taskUUID": "35b9427c-7f64-45ef-94fa-325cadc9b243",
  "audioUUID": "612d211d-ac32-46c9-bcad-f6b551801e4d",
  "audioURL": "https://am.runware.ai/audio/os/a23d05/ws/5/ai/612d211d-ac32-46c9-bcad-f6b551801e4d.mp3",
  "seed": 2213,
  "cost": 0.0945
}
```

--------------------------------

### Text to Video Example

Source: https://runware.ai/docs/models/alibaba-wan2-7/examples

Generates a video from a text prompt using the Wan2.7 model. This example demonstrates the parameters for text-to-video generation, including prompts, dimensions, duration, and optional audio settings.

```APIDOC
## Text to Video

### Description
Generates a video from a text prompt. This operation supports detailed positive and negative prompts, video dimensions, duration, and audio conditioning.

### Method
POST

### Endpoint
/video/inference

### Parameters
#### Request Body
- **taskType** (string) - Required - Must be "videoInference".
- **taskUUID** (string) - Required - Unique identifier for the task.
- **model** (string) - Required - The model to use, e.g., "alibaba:wan@2.7".
- **positivePrompt** (string) - Required - Detailed description of the desired video content.
- **negativePrompt** (string) - Optional - Description of elements to exclude from the video.
- **width** (integer) - Required - The width of the output video in pixels.
- **height** (integer) - Required - The height of the output video in pixels.
- **duration** (integer) - Required - The duration of the video in seconds.
- **seed** (integer) - Optional - Seed for reproducible results.
- **settings** (object) - Optional - Additional settings.
  - **audio** (boolean) - Optional - Whether to generate audio for the video.

### Request Example
```json
{
  "taskType": "videoInference",
  "taskUUID": "08a7bd31-232a-45a4-947f-cd12da384c13",
  "model": "alibaba:wan@2.7",
  "positivePrompt": "A vast glass conservatory built into a rugged seaside cliff at golden afternoon, filled with towering citrus trees, hanging orchids, brass irrigation pipes, and narrow stone walkways over koi pools. The video begins outside with waves striking dark rocks far below, then glides through broken archways into the warm greenhouse air where gardeners in slate-blue coats move carefully among giant leaves. Sunbeams refract through curved glass panels, casting shifting prismatic light across wet stone. A flock of white swifts circles through the upper rafters. Midway, the camera passes a child releasing a small mechanical glider that catches an updraft and sails through the central atrium. End on a wide view of the entire cliffside structure against the sea, with drifting mist, believable scale, natural motion, atmospheric depth, highly detailed textures, elegant cinematic realism.",
  "negativePrompt": "low detail, blurry, distorted anatomy, extra limbs, jittery motion, flicker, text, watermark, logo, oversaturated colors, chaotic composition",
  "width": 1920,
  "height": 1080,
  "duration": 8,
  "seed": 55802,
  "settings": {
    "audio": true
  }
}
```

### Response
#### Success Response (200)
- **taskType** (string) - The type of task performed.
- **taskUUID** (string) - The unique identifier for the task.
- **videoUUID** (string) - The unique identifier for the generated video.
- **videoURL** (string) - The URL to access the generated video.
- **seed** (integer) - The seed used for generation.
- **cost** (number) - The cost associated with the generation.

#### Response Example
```json
{
  "taskType": "videoInference",
  "taskUUID": "08a7bd31-232a-45a4-947f-cd12da384c13",
  "videoUUID": "2eb904c2-9f91-43b4-a1f0-8442299b9bd7",
  "videoURL": "https://vm.runware.ai/video/os/a21d05/ws/5/vi/2eb904c2-9f91-43b4-a1f0-8442299b9bd7.mp4",
  "seed": 55802,
  "cost": 1.2051
}
```
```

--------------------------------

### Museum Heist Audio Guide Request

Source: https://runware.ai/docs/models/alibaba-qwen3-tts-1-7b-customvoice.md

Example request to generate an audio guide with a suspenseful tone. Specifies task type, model, positive prompt for voice style, and detailed speech content with voice and speed settings.

```json
{
  "taskType": "audioInference",
  "taskUUID": "4a9082fd-1865-4261-a0bc-c6cfe96001fc",
  "model": "alibaba:qwen@3-tts-1.7b-customvoice",
  "positivePrompt": "Calm, polished museum guide voice with refined diction and a growing undercurrent of suspense. Begin warmly and informative, then gradually sharpen the tension without becoming frantic.",
  "speech": {
    "text": "Welcome to Gallery Twelve, home of the Harrow Collection. Please pause in front of the porcelain crane with cobalt feathers and a gold-tipped beak. At first glance, it appears delicate, almost weightless, yet the base is reinforced with hidden iron pins, a technique used to protect objects that were expected to travel long distances. If you look closely at the label, you may notice a discrepancy. The accession number ends in 314. The crane before you ends in 341. That is not a printing error. Security has asked that visitors remain where they are and avoid the east corridor. A replacement has been placed on display within the last seven minutes. The original artifact is missing. If you are wearing a gray wool hat, carrying a folded map, and standing nearest the skylight, this message is for you: the case you opened triggered a silent pressure alarm. Please set the object down carefully. Museum staff are approaching, and no one needs to get hurt. For all other guests, thank you for your patience. The tour will resume shortly.",
    "voice": "vivian",
    "language": "English",
    "speed": 0.97
  },
  "settings": {
    "maxNewTokens": 2048
  }
}
```

--------------------------------

### Image to Video Request Example

Source: https://runware.ai/docs/models/klingai-video-o3-4k/examples

This example shows a request to generate a video from an image using the Kling VIDEO O3 4K model. It includes a detailed positive prompt, duration, and specific input frame images.

```json
{
  "taskType": "videoInference",
  "taskUUID": "5040ff19-e635-454d-a29e-10f14ded542e",
  "model": "klingai:kling-video@o3-4k",
  "positivePrompt": "A cinematic high-speed journey across a black volcanic glass desert. Preserve the same lone courier and sleek sand-skimmer from the guided frames. The camera begins low and wide, skimming just above the reflective terrain as the vehicle cuts through dust and ash trails, heat shimmer rippling across the ground. Jagged obsidian formations streak past, steam vents pulse from cracks, and the light shifts from warm sunset into fiery twilight. Emphasize smooth forward momentum, realistic physics, dynamic parallax, detailed reflections in the glassy ground, and a dramatic approach toward a massive glowing fissure. Epic science-fantasy tone, premium 4K detail, strong environmental scale, natural cinematic motion, immersive native sound with engine whine, gravel spray, distant volcanic rumble, and rushing air.",
  "duration": 8,
  "providerSettings": {
    "klingai": {
      "sound": true
    }
  },
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/2728a0b8-2996-42b9-b5f5-406b14f16cc8.jpg",
        "frame": "first"
      },
      {
        "image": "https://assets.runware.ai/assets/inputs/d4692143-e253-4e48-bba1-cd3bab53bc40.jpg",
        "frame": "last"
      }
    ]
  }
}
```

--------------------------------

### Image to Video Generation (10s Duration)

Source: https://runware.ai/docs/models/klingai-1-5-pro/examples

This example shows generating a 10-second video from two input images with specific parameters. Note the CFGScale value.

```json
{
  "taskType": "videoInference",
  "taskUUID": "1848d0b3-acf3-4029-b97a-e414856a39de",
  "model": "klingai:2@2",
  "width": 1920,
  "height": 1080,
  "duration": 10,
  "CFGScale": 0.64,
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/d9dd87ba-89f3-4190-8f6e-c07ba71f23eb.jpg",
        "frame": "first"
      },
      {
        "image": "https://assets.runware.ai/assets/inputs/aada2cdc-e92a-4a6f-a113-8eea72294f7a.jpg",
        "frame": "last"
      }
    ]
  }
}
```

--------------------------------

### ControlNet Preprocess Shuffle Response Example

Source: https://runware.ai/docs/models/controlnet-preprocess-shuffle/examples

This snippet illustrates a successful response from the ControlNet Preprocess Shuffle model. It includes the task type, task UUID, a URL for the generated guide image, cost, and UUIDs for the guide and input images.

```json
{
  "taskType": "imageControlNetPreProcess",
  "taskUUID": "83e84e89-9866-4629-baf0-fd56b45367de",
  "guideImageURL": "https://im.runware.ai/image/os/a06dlim3/ws/3/ii/42c9240a-a16e-4f84-a7a0-1ca0f03c4424.jpg",
  "cost": 0.0006,
  "guideImageUUID": "42c9240a-a16e-4f84-a7a0-1ca0f03c4424",
  "inputImageUUID": "d39c946a-e4c5-4998-95c4-caa1a81e7604"
}
```

--------------------------------

### Example Request and Response

Source: https://runware.ai/docs/models/mediapipe-face-short.md

This snippet provides a full example of a request to the Mediapipe Face Short model and its corresponding response, illustrating how to use the model for image masking.

```APIDOC
## Examples

### Candid Bakery Prep Frame (Mask)

**Request**:

```json
{
  "taskType": "imageMasking",
  "taskUUID": "dd56940b-cc2b-411e-9cd4-af69373703d4",
  "model": "runware:35@7",
  "settings": {
    "confidence": 0.6,
    "maxDetections": 1,
    "maskPadding": 18,
    "maskBlur": 8
  },
  "inputs": {
    "image": "https://assets.runware.ai/assets/inputs/43228de3-125f-4bef-a825-7c05e66725a0.jpg"
  }
}
```

**Response**:

```json
{
  "taskType": "imageMasking",
  "taskUUID": "dd56940b-cc2b-411e-9cd4-af69373703d4",
  "maskImageURL": "https://im.runware.ai/image/os/a02d21/ws/4/ii/731c7cf1-e38d-4076-9099-02244da71843.jpg",
  "cost": 0.0006,
  "maskImageUUID": "731c7cf1-e38d-4076-9099-02244da71843"
}
```
```

--------------------------------

### ControlNet Configuration

Source: https://runware.ai/docs/models/sd-2-1/schema.json

Defines parameters for ControlNet influence, including start and end steps or percentages, and the required model and guide image.

```APIDOC
## ControlNet Configuration

### Description
Configuration for ControlNet image-to-image generation, allowing precise control over when the influence begins and ends.

### Parameters
#### Request Body
- **model** (string) - Required - Identifier for the ControlNet model.
- **guideImage** (string) - Required - The image to guide the ControlNet process (UUID, URL, Data URI, or Base64).
- **endStep** (integer) - Optional - Absolute step number to end ControlNet influence. Must be greater than `startStep` and less than or equal to `steps`.
- **endStepPercentage** (integer) - Optional - Percentage of steps to end ControlNet influence. Must be greater than `startStepPercentage`.
- **startStep** (integer) - Optional - Absolute step number to start ControlNet influence. Must be less than `endStep`.
- **startStepPercentage** (integer) - Optional - Percentage of steps to start ControlNet influence. Must be less than `endStepPercentage`.

### Constraints
- Either `startStep` and `endStep` must be provided, or `startStepPercentage` and `endStepPercentage` must be provided. They cannot be mixed.
- `startStep` must be less than `endStep`.
- `startStepPercentage` must be less than `endStepPercentage`.
- `startStepPercentage` must be between 0 and 99.
- `endStepPercentage` must be between 1 and 100.
```

--------------------------------

### Image to Video Generation Example

Source: https://runware.ai/docs/models/klingai-2-1-standard/examples

Generates a video from a provided image and a detailed positive prompt. This is useful for creating cinematic scenes based on a starting visual.

```json
{
  "taskType": "videoInference",
  "taskUUID": "4923cad8-a610-4604-9846-08e843a33e8d",
  "model": "klingai:5@1",
  "positivePrompt": "Using the provided first-frame image as the opening composition, generate a cinematic video drifting slowly forward through monumental salt-crusted ruins beside a calm sea. Hairline streams of water weave through the white ground, shimmering reflections ripple softly, distant fish flash beneath the clear channels, seabirds wheel far above, and the traveler takes a few careful steps toward the vast arches. Fine salt dust lifts in gentle gusts, fabric edges flutter, light grows warmer across the carved stone and giant shell details, and the camera movement remains smooth, elegant, and grounded in realistic motion. Surreal coastal majesty, luminous atmosphere, high visual fidelity, coherent geometry, rich texture, polished cinematic fantasy.",
  "duration": 5,
  "CFGScale": 0.61,
  "inputs": {
    "frameImages": [
      {
        "image": "https://assets.runware.ai/assets/inputs/4cac0aaf-d7da-443a-a196-f88626f62e87.jpg",
        "frame": "first"
      }
    ]
  }
}
```

```json
{
  "taskType": "videoInference",
  "taskUUID": "4923cad8-a610-4604-9846-08e843a33e8d",
  "videoUUID": "0f8ab0bc-270f-44e7-af72-9aede76fbfbb",
  "videoURL": "https://vm.runware.ai/video/os/a21d05/ws/5/vi/0f8ab0bc-270f-44e7-af72-9aede76fbfbb.mp4",
  "seed": 451885574,
  "cost": 0.1848
}
```