Extensions/RunningHub
ComfyUI Extension

RunningHub

ComfyUI custom nodes for RunningHub cloud workflow execution and AI model invocation

By liangzheng1128·Created 3 months ago·Updated 3 months ago· 0
liangzheng1128/ComfyUI-RunningHub
Nodes20
On cloudLocal install
CategoryRunningHub/Utils, RunningHub
Stars0
Updated3 months ago

Nodes (20)

RH Any to String

The RH Any to String node is a passthrough wearing a converter's hat

RunningHub/Utils
RH Batch Images

Pick and reorder frames with an index string

RunningHub/Utils
RH Download Image

RH Download Image drags any image URL into your graph as a real IMAGE tensor

RunningHub
RH Download Video

RH Download Video pulls a cloud-rendered clip down and can even give you the frames

RunningHub
RH Execute AI App

RH Execute AI App runs a packaged RunningHub app, not a raw workflow

RunningHub
RH Execute Workflow

RH Execute Workflow runs a whole RunningHub job and drags the results home

RunningHub
RH Extract Image

Pull one image out of a batch by index

RunningHub/Utils
RH Image to Image

RH Image to Image sends your local image to a cloud workflow and gets a new one back

RunningHub/Model
RH Image to Video

RH Image to Video animates an image you already have — on someone else's GPU

RunningHub/Model
RH Node Info

RH Node Info is how you inject local values into a cloud workflow

RunningHub
RH Settings

RH Settings is the front door to RunningHub from inside your graph

RunningHub
RH Split Output URLs

RH Split Output URLs turns one messy text block into five clean sockets

RunningHub/Utils
RH Task Cancel

RH Task Cancel is the eject button for a cloud job that's going nowhere

RunningHub
RH Task Outputs

RH Task Outputs grabs the results of a cloud job you already know the ID for

RunningHub
RH Task Status

RH Task Status is the 'is it done yet?' node for cloud jobs

RunningHub
RH Text to Audio

RH Text to Audio is the pack's one voice — a cloud TTS node that returns real audio

RunningHub/Model
RH Text to Image

A cloud txt2img node with a hardcoded agenda

RunningHub/Model
RH Text to Video

A cloud text-to-video node that returns a URL, not frames

RunningHub/Model
RH Upload File

Upload any file type to RunningHub

RunningHub
RH Upload Image

Feed a local image to a cloud workflow — RH Upload Image

RunningHub
Readme

ComfyUI RunningHub Integration

A ComfyUI custom node library for RunningHub — the cloud-based AI content creation platform. Execute workflows, invoke AI models, and manage cloud resources directly from your local ComfyUI instance.

Features

  • 20 custom nodes covering workflow execution, AI model invocation, file management, and utilities
  • WebSocket real-time progress tracking with HTTP polling fallback
  • Dual install support — git clone or pip install
  • Compatible with both RunningHub China (runninghub.cn) and International (runninghub.ai)

Installation

Method 1: Git Clone (Recommended)

cd ComfyUI/custom_nodes/
git clone https://github.com/YOUR_USERNAME/ComfyUI-RunningHub.git
cd ComfyUI-RunningHub
pip install -r requirements.txt

Method 2: pip install

pip install ComfyUI-RunningHub

Restart ComfyUI after installation.

Quick Start

  1. Get your API Key: Log in to RunningHub → User Menu → API Key
  2. Create a Workflow: Build a workflow on RunningHub and copy its Workflow ID from the URL
  3. Configure: Add an RH Settings node and paste your API Key and Workflow ID
  4. Execute: Add an RH Execute Workflow node, connect Settings → Execute, and queue the prompt

Nodes

Configuration

| Node | Description | |------|-------------| | RH Settings | Configure API Key, Base URL, and Workflow ID | | RH Node Info | Build parameter overrides for workflow nodes |

Task Execution

| Node | Description | |------|-------------| | RH Execute Workflow | Submit and execute a ComfyUI workflow on RunningHub | | RH Execute AI App | Submit and execute an AI App | | RH Task Status | Query task execution status | | RH Task Cancel | Cancel a running task | | RH Task Outputs | Retrieve task output results |

AI Model Invocation

| Node | Description | |------|-------------| | RH Text to Image | Generate images from text | | RH Image to Image | Transform images with a prompt | | RH Image to Video | Generate video from an image | | RH Text to Video | Generate video from text | | RH Text to Audio | Generate audio from text |

File Management

| Node | Description | |------|-------------| | RH Upload Image | Upload an IMAGE tensor to RunningHub | | RH Upload File | Upload a local file to RunningHub | | RH Download Image | Download a URL as IMAGE tensor | | RH Download Video | Download a URL as video file |

Utilities

| Node | Description | |------|-------------| | RH Any to String | Convert any value to string | | RH Extract Image | Extract a single image from a batch | | RH Batch Images | Select images by index range | | RH Split Output URLs | Split multi-line URLs into individual outputs |

Typical Workflow

RH Settings → RH Node Info → RH Execute Workflow → RH Download Image
  1. RH Settings: Provide your API Key and Workflow ID
  2. RH Node Info: Set prompt text, seeds, or other parameters (chain multiple for multi-param)
  3. RH Execute Workflow: Submit the task and wait for results
  4. RH Download Image: Convert output URLs to ComfyUI IMAGE tensors for further processing

Configuration

Base URLs

  • China: https://www.runninghub.cn
  • International: https://www.runninghub.ai

GPU Options

The Execute nodes support an optional use_rtx4090 flag for high-performance GPU execution.

API Reference

RunningHub API documentation:

Development

# Install in development mode
pip install -e ".[dev]"

# Run tests
pytest tests/ -v

# Build package
python -m build

License

MIT License. See LICENSE for details.