Runway

Access Runway's AI models including Gen-4 Turbo and Aleph for video generation through Runware's unified API. Learn about Runway-specific parameters and cinematic video capabilities.

Introduction

Runway's AI models are integrated into the Runware platform through our unified API, providing access to cinematic video generation technology built for professional storytelling and creative iteration.

Through the providerSettings.runway object, you can access Runway's content moderation controls while maintaining the consistency of Runware's standard API structure. This page documents the technical specifications, parameter requirements, and provider-specific settings for all Runway models available through our platform.

Image models

Runway Gen-4 Image

Runway's Gen-4 Image is a high-fidelity text-to-image model offering advanced stylistic control and visual consistency across scenes and characters. Built for professional creative workflows, it delivers exceptional detail and artistic precision with optional reference image support.

Model AIR ID: runway:4@1.

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

Technical specifications:

  • Positive prompt: 1-1000 characters (required).
  • Reference images: Supports up to 3 images via inputs.referenceImages (optional).
  • Reference tags: Use @tag syntax in prompts to reference specific images. Tags must be 3-16 alphanumeric characters starting with a letter.
  • Supported dimensions: 1920×1080 (16:9), 1080×1920 (9:16), 1024×1024 (1:1), 1360×768 (~16:9), 1080×1080 (1:1), 1168×880 (~4:3), 1440×1080 (4:3), 1080×1440 (3:4), 1808×768 (~21:9), 2112×912 (~21:9), 1280×720 (16:9), 720×1280 (9:16), 720×720 (1:1), 960×720 (4:3), 720×960 (3:4), 1680×720 (~21:9).

Provider-specific settings:

Parameters supported: contentModeration.

{
  "taskType": "imageInference",
  "taskUUID": "f47ac10b-58cc-4372-a567-0e02b2c3d493",
  "model": "runway:4@1",
  "positivePrompt": "A cinematic portrait with dramatic lighting and rich color palette",
  "width": 1920,
  "height": 1080
}
{
  "taskType": "imageInference",
  "taskUUID": "6ba7b831-9dad-11d1-80b4-00c04fd430c8",
  "model": "runway:4@1",
  "inputs": {
    "referenceImages": [
      {
        "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
        "tag": "character"
      },
      {
        "image": "d7e8f9a0-2b5c-4e7f-a1d3-9c8b7a6e5d4f",
        "tag": "style"
      }
    ]
  },
  "positivePrompt": "Create a scene with @character in the artistic style of @style",
  "width": 1440,
  "height": 1080
}
{
  "taskType": "imageInference",
  "taskUUID": "550e8400-e29b-41d4-a716-446655440012",
  "model": "runway:4@1",
  "positivePrompt": "A professional portrait with natural lighting",
  "width": 1024,
  "height": 1024,
  "providerSettings": {
    "runway": {
      "contentModeration": {
        "publicFigureThreshold": "low"
      }
    }
  }
}

Runway Gen-4 Image Turbo

Runway's Gen-4 Image Turbo is the faster variant focused on rapid iterations and quick concept generation while preserving key visual quality and control. Optimized for creative experimentation and workflow velocity, it requires reference images to guide generation.

Model AIR ID: runway:4@2.

Supported workflows: Image-to-image.

Technical specifications:

  • Positive prompt: 1-1000 characters (required).
  • Reference images: Required via inputs.referenceImages (1-3 images).
  • Reference tags: Use @tag syntax in prompts to reference specific images. Tags must be 3-16 alphanumeric characters starting with a letter.
  • Supported dimensions: 1920×1080 (16:9), 1080×1920 (9:16), 1024×1024 (1:1), 1360×768 (~16:9), 1080×1080 (1:1), 1168×880 (~4:3), 1440×1080 (4:3), 1080×1440 (3:4), 1808×768 (~21:9), 2112×912 (~21:9), 1280×720 (16:9), 720×1280 (9:16), 720×720 (1:1), 960×720 (4:3), 720×960 (3:4), 1680×720 (~21:9).

Provider-specific settings:

Parameters supported: contentModeration.publicFigureThreshold.

{
  "taskType": "imageInference",
  "taskUUID": "f47ac10b-58cc-4372-a567-0e02b2c3d494",
  "model": "runway:4@2",
  "inputs": {
    "referenceImages": [
      {
        "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
        "tag": "subject"
      }
    ]
  },
  "positivePrompt": "Create a dynamic composition featuring @subject with bold colors",
  "width": 1920,
  "height": 1080
}
{
  "taskType": "imageInference",
  "taskUUID": "6ba7b832-9dad-11d1-80b4-00c04fd430c8",
  "model": "runway:4@2",
  "inputs": {
    "referenceImages": [
      {
        "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
        "tag": "character"
      },
      {
        "image": "d7e8f9a0-2b5c-4e7f-a1d3-9c8b7a6e5d4f",
        "tag": "environment"
      },
      {
        "image": "e8f9a0b1-3c6d-5e8f-b2d4-0d1e2f3a4b5c",
        "tag": "lighting"
      }
    ]
  },
  "positivePrompt": "Place @character in @environment with the mood of @lighting",
  "width": 2112,
  "height": 912
}
{
  "taskType": "imageInference",
  "taskUUID": "550e8400-e29b-41d4-a716-446655440013",
  "model": "runway:4@2",
  "inputs": {
    "referenceImages": [
      {
        "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
        "tag": "ref"
      }
    ]
  },
  "positivePrompt": "Transform @ref into a futuristic cityscape with neon lighting",
  "width": 1680,
  "height": 720
}

Video models

providerSettings » runway
runway
object

Configuration object for Runway-specific settings and controls.

View example 
"providerSettings": {
  "runway": { 
    "contentModeration": { 
      "publicFigureThreshold": "low"
    } 
  } 
}
Properties ⁨1⁩ property
providerSettings » runway » contentModeration
contentModeration
object

Configuration object for content moderation settings. Controls how strictly the model filters potentially sensitive content, particularly regarding recognizable public figures.

Object properties:

  • publicFigureThreshold (string, optional, default: "auto"): Controls how strictly the model prevents generation of recognizable public figures in the output video.

Available values for publicFigureThreshold:

  • auto: Automatically determines the appropriate threshold based on content.
  • low: Less strict filtering, may allow more recognizable public figures.
View example 
"providerSettings": {
  "runway": {
    "contentModeration": {
      "publicFigureThreshold": "low"
    }
  }
}

Runway Gen-4 Turbo

Runway’s Gen-4 Turbo is a high-speed image-to-video model built for instant creative experimentation. It produces dynamic, cinematic motion with exceptional fluidity and detail, transforming a single image into a fully realized video in seconds.

Model AIR ID: runway:1@1.

Supported workflows: Image-to-video.

Technical specifications:

  • Positive prompt: 1-1000 characters (optional).
  • Supported dimensions: 1280×720 (16:9), 720×1280 (9:16), 1104×832, 832×1104, 960×960 (1:1), 1584×672 (21:9).
  • Duration: 2-10 seconds (default: 10).
  • Frame images: Supports first frame for inputs.frameImages (required).
  • Input image requirements: 20MB file size limit.

Provider-specific settings:

Parameters supported: contentModeration.publicFigureThreshold.

{
  "taskType": "videoInference",
  "taskUUID": "6ba7b829-9dad-11d1-80b4-00c04fd430c8",
  "model": "runway:1@1",
  "inputs": {
    "frameImages": [
      {
        "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
        "frame": "first"
      }
    ],
    "positivePrompt": "The scene comes to life with dynamic movement",
    "duration": 10,
    "width": 1280,
    "height": 720
  }
}
{
  "taskType": "videoInference",
  "taskUUID": "550e8400-e29b-41d4-a716-446655440010",
  "model": "runway:1@1",
  "positivePrompt": "A portrait shot with natural lighting",
  "duration": 5,
  "width": 960,
  "height": 960,
  "providerSettings": {
    "runway": {
      "contentModeration": {
        "publicFigureThreshold": "low"
      }
    }
  }
}

Runway Aleph

Runway's flagship video-to-video model is built for realism and expressive storytelling. It transforms videos with cinematic precision and supports one reference image to guide visual style or content direction.

Model AIR ID: runway:2@1.

Supported workflows: Video-to-video.

Technical specifications:

  • Positive prompt: 1-1000 characters (required).
  • Input video: Supports inputs.video (required).
  • Reference images: Supports inputs.referenceImages with 1 image (optional).
  • Supported dimensions: 1280×720 (16:9), 720×1280 (9:16), 1104×832, 832×1104, 960×960 (1:1), 1584×672 (21:9), 848×480, 640×480.
  • Input video requirements: 20MB file size limit.
  • Input image requirements: 20MB file size limit.

Provider-specific settings:

Parameters supported: contentModeration.publicFigureThreshold.

{
  "taskType": "videoInference",
  "taskUUID": "550e8400-e29b-41d4-a716-446655440011",
  "model": "runway:2@1",
  "inputs": {
    "video": "d7e8f9a0-2b5c-4e7f-a1d3-9c8b7a6e5d4f"
  },
  "positivePrompt": "Enhance this video with cinematic color grading and smooth motion",
  "width": 1920,
  "height": 1080
}
{
  "taskType": "videoInference",
  "taskUUID": "6ba7b830-9dad-11d1-80b4-00c04fd430c8",
  "model": "runway:2@1",
  "inputs": {
    "video": "d7e8f9a0-2b5c-4e7f-a1d3-9c8b7a6e5d4f"
    "referenceImages": ["c64351d5-4c59-42f7-95e1-eace013eddab"]
  },
  "positivePrompt": "Transform this scene with dynamic camera movement and atmospheric lighting",
  "width": 1280,
  "height": 720
}
Ask AI
Context: Full page
Send feedback
Context: Full page