Google

Access Google's AI models including Imagen for image generation and Veo for video creation through Runware's unified API. Learn about Google-specific parameters, limitations, and technical specifications.

Introduction

Google's AI models are integrated into the Runware platform through our unified API, providing access to advanced image and video generation technology.

Through the providerSettings.google object, you can access Google's unique features such as automatic prompt enhancement and native audio generation, while maintaining the consistency of Runware's standard API structure. This page documents the technical specifications, parameter requirements, and provider-specific settings for all Google models available through our platform.

providerSettings » google
google
object

Configuration settings specific to Google's video generation models (Veo 2 and Veo 3). These settings control various aspects of the generation process including prompt enhancement and audio generation capabilities.

View example 
"providerSettings": {
  "google": { 
    "enhancePrompt": true,
    "generateAudio": false
  } 
}
Properties ⁨2⁩ properties
providerSettings » google » enhanceprompt
enhanceprompt
boolean Default: true

Controls whether the input prompt is automatically enhanced and expanded to improve generation quality. When enabled, the system optimizes the prompt for better results by adding relevant details and context.

This setting cannot be disabled when using Veo 3 model, as prompt enhancement is always active. For Veo 2 model, this setting can be controlled and disabled if needed.

Enhanced prompts typically result in more detailed and higher-quality video generation by providing the model with richer context and clearer instructions.

When prompt enhancement is enabled, reproducibility is not guaranteed even when using the same seed value. The enhancement process may introduce variability that affects the deterministic nature of generation.

providerSettings » google » generateaudio
generateaudio
boolean Default: false

Controls whether the generated video includes audio content. When enabled, the system creates appropriate audio that matches the visual content and scene context within the video.

This feature is only available for Veo 3 model. Audio generation is not supported in Veo 2.

Generated audio can include ambient sounds, music, or other audio elements that enhance the video experience and provide a more immersive result.

Image models

Imagen 3.0

Imagen 3.0 creates detailed, high-quality images with better lighting and fewer artifacts. It works well for both realistic scenes and stylized visuals.

Model AIR ID: google:1@1.

Supported workflows: Text-to-image.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Supported dimensions: 1024×1024 (1:1), 768×1408 (9:16), 1408×768 (16:9), 896×1280 (3:4), 1280×896 (4:3).
  • Output format: JPG, PNG, WEBP.
{
  "taskType": "imageInference",
  "taskUUID": "24cd5dff-cb81-4db5-8506-b72a9425f9d1",
  "model": "google:1@1",
  "positivePrompt": "A detailed landscape photograph of a serene mountain lake at sunrise with mist rising from the water",
  "width": 1408,
  "height": 768
}

Imagen 3.0 Fast

Imagen 3.0 Fast is a quicker version of Imagen 3, built for speed without sacrificing much quality—ideal for interactive or real-time use.

Model AIR ID: google:1@2.

Supported workflows: Text-to-image.

Technical specifications:

  • Positive prompt 2-3000 characters.
  • Negative prompt: 2-3000 characters (optional).
  • Supported dimensions: 1024×1024 (1:1), 768×1408 (9:16), 1408×768 (16:9), 896×1280 (3:4), 1280×896 (4:3).
{
  "taskType": "imageInference",
  "taskUUID": "b8c4d952-7f27-4a6e-bc9a-83f01d1c6d59",
  "model": "google:1@2",
  "positivePrompt": "A modern architectural building with clean geometric lines and glass facades",
  "negativePrompt": "blurry, low quality, distorted, unrealistic",
  "width": 896,
  "height": 1280
}

Imagen 4.0 Preview

Imagen 4.0 Preview improves textures, lighting, and typography—making it especially useful for design-heavy or detail-focused work.

Model AIR ID: google:2@1.

Supported workflows: Text-to-image.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Supported dimensions: 1024×1024 (1:1), 768×1408 (9:16), 1408×768 (16:9), 896×1280 (3:4), 1280×896 (4:3).
{
  "taskType": "imageInference",
  "taskUUID": "f3a2b8c9-1e47-4d3a-9b2f-8c7e6d5a4b3c",
  "model": "google:2@1",
  "positivePrompt": "A photorealistic portrait of an elderly craftsman working with traditional woodworking tools in his workshop",
  "width": 1024,
  "height": 1024
}

Imagen 4.0 Ultra

Imagen 4.0 Ultra is Google's most advanced image model available, delivering exceptional detail, color accuracy, and prompt adherence. Ideal for demanding use cases where image quality and consistency matter most.

Model AIR ID: google:2@2.

Supported workflows: Text-to-image.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Supported dimensions: 1024×1024 (1:1), 768×1408 (9:16), 1408×768 (16:9), 896×1280 (3:4), 1280×896 (4:3).
{
  "taskType": "imageInference",
  "taskUUID": "e4d3c2b1-5a6f-4c8e-b2d7-1f0e9d8c7b6a",
  "model": "google:2@2",
  "positivePrompt": "An ultra-detailed macro photograph of a butterfly wing showing intricate patterns and iridescent colors with perfect focus and lighting",
  "width": 1280,
  "height": 896
}

Imagen 4.0 Fast

Imagen 4.0 Fast offers the speed and quality of the Imagen 4 family, optimized for quicker inference with minimal quality loss. It’s suited for fast generation tasks without giving up fine detail or lighting accuracy.

Model AIR ID: google:2@3.

Supported workflows: Text-to-image.

Technical specifications:

  • Positive prompt 2-3000 characters.
  • Negative prompt: 2-3000 characters (optional).
  • Supported dimensions: 1024×1024 (1:1), 768×1408 (9:16), 1408×768 (16:9), 896×1280 (3:4), 1280×896 (4:3).
{
  "taskType": "imageInference",
  "taskUUID": "1a2b3c4d-5e6f-7a8b-9c0d-123456789abc",
  "model": "google:2@3",
  "positivePrompt": "A vibrant cityscape at golden hour with warm lighting and dynamic reflections",
  "negativePrompt": "low detail, blurry, overexposed",
  "width": 1408,
  "height": 768
}

Video models

Veo 2

Google's Veo 2 model delivers high-realism video generation with advanced physics simulation and cinematic understanding, supporting both text-to-video and image-to-video generation with automatic prompt enhancement.

Model AIR ID: google:2@0.

Supported workflows: Text-to-video, image-to-video.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Supported dimensions: 1280×720 (16:9), 720×1280 (9:16).
  • Frame rate: 24 FPS.
  • Duration: 5, 6, 7, or 8 seconds (default: 5).
  • Frame images: Supports first and last frame for frameImages.
  • Input image requirements: Width and height between 300-2048 pixels, 20MB file size limit.

Provider-specific parameters supported: enhancePrompt.

{
  "taskType": "videoInference",
  "taskUUID": "24cd5dff-cb81-4db5-8506-b72a9425f9d1",
  "model": "google:2@0",
  "positivePrompt": "A close-up shot of rain drops on a window",
  "duration": 6,
  "width": 1280,
  "height": 720
}
{
  "taskType": "videoInference",
  "taskUUID": "b8c4d952-7f27-4a6e-bc9a-83f01d1c6d59",
  "model": "google:2@0",
  "frameImages": [
    {
      "inputImage": "c64351d5-4c59-42f7-95e1-eace013eddab",
      "frame": "first"
    }
  ],
  "duration": 8,
  "width": 720,
  "height": 1280,
  "providerSettings": {
    "google": {
      "enhancePrompt": false
    }
  }
}

Veo 3

Google's Veo 3 model represents Google's latest video generation technology, featuring native audio generation that creates synchronized dialogue, music, and sound effects alongside high-fidelity video content.

Model AIR ID: google:3@0.

Supported workflows: Text-to-video, image-to-video.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Supported dimensions: 1280×720 (16:9).
  • Frame rate: 24 FPS.
  • Duration: 8 seconds.
  • Frame images: Supports first frame for frameImages.
  • Input image requirements: Width and height between 300-2048 pixels, 20MB file size limit.

Provider-specific parameters supported: generateAudio.

In Veo 3, enhancePrompt is always enabled and cannot be disabled.

{
  "taskType": "videoInference",
  "taskUUID": "f3a2b8c9-1e47-4d3a-9b2f-8c7e6d5a4b3c",
  "model": "google:3@0",
  "positivePrompt": "Ocean waves crashing against rocky cliffs during a storm",
  "duration": 8,
  "width": 1280,
  "height": 720,
  "providerSettings": {
    "google": {
      "generateAudio": true
    }
  }
}
{
  "taskType": "videoInference",
  "taskUUID": "e4d3c2b1-5a6f-4c8e-b2d7-1f0e9d8c7b6a",
  "model": "google:3@0",
  "positivePrompt": "A time-lapse of clouds moving across a mountain landscape",
  "duration": 8,
  "width": 1280,
  "height": 720,
  "providerSettings": {
    "google": {
      "generateAudio": false
    }
  }
}

Veo 3 Fast

Google's Veo 3 Fast model represents a faster and more cost-effective variant of Veo 3, optimized for speed and affordability while maintaining native audio generation capabilities.

Model AIR ID: google:3@1.

Supported workflows: Text-to-video, image-to-video.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Supported dimensions: 1280×720 (16:9).
  • Frame rate: 24 FPS.
  • Duration: 8 seconds.
  • Frame images: Supports first frame for frameImages.
  • Input image requirements: Width and height between 300-2048 pixels, 20MB file size limit.

Provider-specific parameters supported: generateAudio.

In Veo 3, enhancePrompt is always enabled and cannot be disabled.

{
  "taskType": "videoInference",
  "taskUUID": "f3a2b8c9-1e47-4d3a-9b2f-8c7e6d5a4b3c",
  "model": "google:3@1",
  "positivePrompt": "Fast-paced street scene with cars and pedestrians, urban ambient sounds",
  "duration": 8,
  "width": 1280,
  "height": 720,
  "providerSettings": {
    "google": {
      "generateAudio": true
    }
  }
}
{
  "taskType": "videoInference",
  "taskUUID": "e4d3c2b1-5a6f-4c8e-b2d7-1f0e9d8c7b6a",
  "model": "google:3@1",
  "positivePrompt": "Quick animation of clouds forming and dispersing over a landscape",
  "duration": 8,
  "width": 1280,
  "height": 720,
  "providerSettings": {
    "google": {
      "generateAudio": false
    }
  }
}