ComfyUI Extension: Seedream4 Replicate

Authored by Saganaki22

Created

Updated

2 stars

ComfyUI node for Seedream 4 image generation via Replicate API with support for text-to-image and image-to-image generation

Custom Nodes (0)

    README

    ComfyUI Seedream 4 Replicate

    Replicate ComfyUI Python

    A ComfyUI custom node for ByteDance's Seedream 4 image generation model via the Replicate API. Generate high-quality images with advanced control over aspect ratios, multi-image generation, and seed management.

    Gallery

    | | | | |:---:|:---:|:---:| | <img width="309" height="335" alt="Einstein Portrait" src="https://github.com/user-attachments/assets/731a09f6-b156-4905-84d8-efd94daec566" /> |<img width="309" height="335" alt="Node Interface" src="https://github.com/user-attachments/assets/ca695ba4-2555-47bf-86c9-366b297a8e1f" /> | <img width="231" height="308" alt="Portrait Example" src="https://github.com/user-attachments/assets/0f764a9d-6716-4228-b41f-7119f22e6ddc" /> |

    Installation

    1. Clone this repository into your ComfyUI custom_nodes folder:
    git clone https://github.com/Saganaki22/ComfyUI-Seedream4_Replicate.git
    1. Install dependencies:
    pip install -r requirements.txt

    1. Get your Replicate API token from replicate.com/account/api-tokens

    2. Restart ComfyUI

    Node Parameters

    | Parameter | Type | Description | |-----------|------|-------------| | API Key | String | Your Replicate API token (starts with r8_) | | Prompt | String | Text description of what you want to generate | | Size Preset | Dropdown | Common aspect ratios (1:1, 4:3, 16:9, etc.) or Custom | | Width | Integer | Custom width in pixels (1024-4096, used with Custom preset) | | Height | Integer | Custom height in pixels (1024-4096, used with Custom preset) | | Max Images | Integer | Maximum images to generate (1-15, only with Sequential mode) | | Sequential Generation | Dropdown | disabled (single image) or auto (let AI decide multiple) | | Seed | Integer | Random seed for reproducible results (-1 for random) | | Control After Generate | Dropdown | Seed behavior after generation (fixed/increment/decrement/randomize) | | Image Input | Image | Optional reference image for image-to-image generation |

    Features

    • Multiple aspect ratios with preset sizes
    • Sequential image generation for related image sets
    • Image-to-image support with reference images
    • Seed control for reproducible results
    • Real-time progress tracking with 4-stage progress updates
    • Automatic cancellation support with ComfyUI's cancel button
    • Error handling with content policy detection

    Usage Tips

    • Use Sequential mode with prompts like "story sequence" or "character variations" for multiple related images
    • Custom dimensions work best for specific use cases outside standard ratios
    • Seed control lets you iterate on prompts while maintaining composition
    • Large input images may exceed data URL limits (shows warning)

    License

    MIT License - see LICENSE file for details

    <p align="center"> <a href="https://replicate.com/" target="_blank"> <img width="150" height="150" alt="image" src="https://github.com/user-attachments/assets/ef0e9c4c-5174-41f4-a956-0d2460a34820" /> </a> </p>