ByteDance

Access ByteDance's AI models including Seedance for video generation through Runware's unified API. Learn about ByteDance-specific parameters, limitations, and technical specifications.

Introduction

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

Through the providerSettings.bytedance object, you can access ByteDance's unique features such as camera movement 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 ByteDance models available through our platform.

providerSettings » bytedance bytedance object

Configuration settings specific to ByteDance's video generation models. These settings control camera behavior and movement during video generation.

View example

"providerSettings": {
  "bytedance": { 
    "cameraFixed": false
  } 
}

Properties ⁨1⁩ property

providerSettings » bytedance » camerafixed camerafixed boolean Default: false

Controls whether the camera remains stationary during video generation. When enabled, the camera position and angle are fixed, preventing any camera movement, panning, or zooming effects.

When disabled (default), the model can incorporate dynamic camera movements such as pans, tilts, zooms, or tracking shots to create more cinematic and engaging video content.

This setting is useful when you need static shots or want to avoid camera motion that might distract from the main subject or action in the video.

Image models

SeedEdit 3.0

ByteDance's SeedEdit 3.0 model provides flagship image editing capabilities with advanced instruction-following and content preservation, balancing prompt fidelity with strong preservation of unedited areas.

Model AIR ID: bytedance:4@1.

Supported workflows: Image-to-image.

Technical specifications:

Positive prompt: 2-500 characters.
Output dimensions: Aspect ratio inherited from reference image (up to 4K resolution).
Reference images: Supports referenceImages with 1 image (required).
CFG Scale: 1-10 (default: 5.5).
Input image requirements: Width and height between 300-6000 pixels, 10MB file size limit.

{
  "taskType": "imageInference",
  "taskUUID": "24cd5dff-cb81-4db5-8506-b72a9425f9d1",
  "model": "bytedance:4@1",
  "positivePrompt": "Change the lighting to warm golden hour, remove the background clutter",
  "referenceImages": [
    "c64351d5-4c59-42f7-95e1-eace013eddab"
  ],
  "CFGScale": 0.7
}

Video models

Seedance 1.0 Lite

ByteDance's Seedance 1.0 Lite model provides cost-effective video generation with support for multiple resolution options across various aspect ratios, supporting both text-to-video and image-to-video workflows.

Model AIR ID: bytedance:1@1.

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

Technical specifications:

Positive prompt: 2-3000 characters.
Supported dimensions: 864×480 (16:9), 736×544 (4:3), 640×640 (1:1), 960×416 (21:9), 416×960 (9:21), 1248×704 (16:9), 1120×832 (4:3), 960×960 (1:1), 1504×640 (21:9), 640×1504 (9:21).
Frame rate: 24 FPS.
Duration: 5 or 10 seconds (default: 5).
Frame images: Supports first and last frame for frameImages.
Input image requirements: Width and height between 300-6000 pixels, 10MB file size limit.

When using frameImages (image-to-video), the width and height parameters are not supported. The output resolution is automatically determined by selecting the closest supported resolution that matches the input image's aspect ratio.

Provider-specific parameters supported: cameraFixed.

Text-to-video

{
  "taskType": "videoInference",
  "taskUUID": "24cd5dff-cb81-4db5-8506-b72a9425f9d1",
  "model": "bytedance:1@1",
  "positivePrompt": "A serene mountain lake reflecting the sunset with gentle ripples",
  "duration": 5,
  "width": 1248,
  "height": 704
}

Image-to-video

{
  "taskType": "videoInference",
  "taskUUID": "b8c4d952-7f27-4a6e-bc9a-83f01d1c6d59",
  "model": "bytedance:1@1",
  "frameImages": [
    {
      "inputImage": "c64351d5-4c59-42f7-95e1-eace013eddab",
      "frame": "first"
    }
  ],
  "duration": 10,
  "providerSettings": {
    "bytedance": {
      "cameraFixed": true
    }
  }
}

Seedance 1.0 Pro

ByteDance's Seedance 1.0 Pro model represents the flagship professional-grade video generation solution with cinematic storytelling capabilities, multi-shot support, and high-resolution output up to 1080p.

Model AIR ID: bytedance:2@1.

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

Technical specifications:

Positive prompt: 2-3000 characters.
Supported dimensions: 864×480 (16:9), 736×544 (4:3), 640×640 (1:1), 544×736 (3:4), 480×864 (9:16), 416×960 (9:21), 960×416 (21:9), 1920×1088 (16:9), 1664×1248 (4:3), 1440×1440 (1:1), 1248×1664 (3:4), 1088×1920 (9:16), 928×2176 (9:21), 2176×928 (21:9).
Frame rate: 24 FPS.
Duration: 5 or 10 seconds (default: 5).
Frame images: Supports first frame for frameImages.
Input image requirements: Width and height between 300-6000 pixels, 10MB file size limit.

Provider-specific settings:

Parameters supported: cameraFixed.

Text-to-video

{
  "taskType": "videoInference",
  "taskUUID": "24cd5dff-cb81-4db5-8506-b72a9425f9d1",
  "model": "bytedance:2@1",
  "positivePrompt": "Cinematic multi-shot sequence: close-up of a coffee cup, camera pulls back to reveal a busy café, then cuts to street view through window",
  "duration": 10,
  "width": 1920,
  "height": 1088,
  "providerSettings": {
    "bytedance": {
      "cameraFixed": false
    }
  }
}

Image-to-video

{
  "taskType": "videoInference",
  "taskUUID": "b8c4d952-7f27-4a6e-bc9a-83f01d1c6d59",
  "model": "bytedance:2@1",
  "frameImages": [
    {
      "inputImage": "c64351d5-4c59-42f7-95e1-eace013eddab",
      "frame": "first"
    }
  ],
  "duration": 5,
  "width": 1440,
  "height": 1440,
  "providerSettings": {
    "bytedance": {
      "cameraFixed": true
    }
  }
}