https://github.com/leonardomiramondi/flux-context-comfyui
ComfyUI custom node for Flux image generation with context support via Replicate API
https://github.com/leonardomiramondi/flux-context-comfyui
Last synced: 1 day ago
JSON representation
ComfyUI custom node for Flux image generation with context support via Replicate API
- Host: GitHub
- URL: https://github.com/leonardomiramondi/flux-context-comfyui
- Owner: leonardomiramondi
- License: mit
- Created: 2025-06-24T09:43:08.000Z (2 days ago)
- Default Branch: main
- Last Pushed: 2025-06-24T10:24:00.000Z (2 days ago)
- Last Synced: 2025-06-24T11:33:31.704Z (2 days ago)
- Language: Python
- Size: 15.6 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-comfyui - **Flux Context ComfyUI Node**
README
# Flux Context ComfyUI Node
A ComfyUI custom node for **Flux Context (Kontext)** - advanced image editing and transformation using text prompts via the Replicate API.
## What is Flux Context?
Flux Context (Kontext) is Black Forest Labs' state-of-the-art **image editing** model that allows you to transform existing images using natural language descriptions. Unlike image generation models, Flux Context specializes in:
- **Style Transfer**: Convert photos to different art styles (watercolor, oil painting, sketches)
- **Object/Clothing Changes**: Modify hairstyles, add accessories, change colors
- **Text Editing**: Replace text in signs, posters, and labels
- **Background Swapping**: Change environments while preserving subjects
- **Character Consistency**: Maintain identity across multiple edits
## Features
- ✅ **Text-based image editing** using natural language prompts
- ✅ **Two image inputs** ("image 1" required, "image 2" optional for reference)
- ✅ **Professional models**: Flux Kontext Pro and Max
- ✅ **API token input** built into the node interface
- ✅ **Multiple output formats** (JPG/PNG)
- ✅ **Smart image compression** for API compatibility
- ✅ **Error handling** with automatic retry logic
## Available Models
- **black-forest-labs/flux-kontext-pro**: State-of-the-art performance with high-quality outputs
- **black-forest-labs/flux-kontext-max**: Premium model with maximum performance and improved typography
## Installation
### Method 1: ComfyUI Manager (Recommended)
1. Install [ComfyUI Manager](https://github.com/ltdrdata/ComfyUI-Manager)
2. Search for "Flux Context" in the manager
3. Install and restart ComfyUI
### Method 2: Manual Installation
1. **Find your ComfyUI installation directory**:
```bash
# Common locations:
# Windows: C:\ComfyUI\ or C:\Users\[username]\ComfyUI\
# Mac: /Applications/ComfyUI/ or ~/ComfyUI/
# Linux: ~/ComfyUI/ or /opt/ComfyUI/
```
2. **Navigate to custom nodes directory**:
```bash
cd [your-comfyui-path]/custom_nodes/
```
3. **Clone this repository**:
```bash
git clone https://github.com/leonardomiramondi/flux-context-comfyui.git
```
4. **Install dependencies**:
```bash
cd flux-context-comfyui
pip install -r requirements.txt
```
5. **Restart ComfyUI** completely
## Setup
1. **Get a Replicate API token**:
- Go to [replicate.com/account/api-tokens](https://replicate.com/account/api-tokens)
- Create a new token
- Copy the token (starts with `r8_`)
2. **Add the node to your workflow**:
- In ComfyUI, add the **Flux Context Node** (found under `image/editing`)
- Paste your API token in the "api_token" field
## Usage
### Basic Image Editing
1. Connect an image to **"image 1"** (required)
2. Enter your **editing prompt** describing the desired changes
3. Choose your **model** (Pro or Max)
4. Select **output format** (JPG or PNG)
5. Run the workflow
### Reference-Style Editing
1. Connect your main image to **"image 1"**
2. Connect a reference/style image to **"image 2"** (optional)
3. Describe the transformation in the **editing prompt**
4. The model will use image 2 as a style reference
### Example Prompts
**Style Transfer:**
- "Transform this into a watercolor painting"
- "Make this look like a Renaissance oil painting"
- "Convert to a pencil sketch with detailed shading"
**Object Changes:**
- "Change the red car to a blue motorcycle"
- "Replace the person's outfit with a business suit"
- "Add sunglasses and a hat to the person"
**Background Edits:**
- "Change the background to a beach scene while keeping the person"
- "Replace the indoor setting with a forest"
- "Add falling snow to this winter scene"
**Text Editing:**
- "Change the sign text from 'OPEN' to 'CLOSED'"
- "Replace the billboard text with 'SALE 50% OFF'"
## API Requirements
- **Replicate account** with available credits
- **Valid API token** (get from [replicate.com](https://replicate.com/account/api-tokens))
- **Internet connection** for API calls
## Troubleshooting
### "Invalid API token"
- Verify your token starts with `r8_`
- Check you have credits in your Replicate account
- Ensure the token has proper permissions
### "Image too large" errors
- The node automatically compresses large images
- Try reducing your input image resolution if issues persist
- Maximum recommended size: 2048x2048 pixels
### Node not appearing
- Ensure you've restarted ComfyUI completely after installation
- Check that `requirements.txt` packages installed successfully
- Verify the node files are in the correct custom_nodes directory
### "Model not found" errors
- Only use the supported models (flux-kontext-pro or flux-kontext-max)
- Ensure your Replicate account has access to these models
## Contributing
Feel free to submit issues, feature requests, or pull requests on the [GitHub repository](https://github.com/leonardomiramondi/flux-context-comfyui).
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## Credits
- **Black Forest Labs** for the Flux Context models
- **Replicate** for API infrastructure
- **ComfyUI** community for the amazing framework