Introducing Redux: A New Tool for Remixing Images and Text Prompts
What is Redux?
Black Forest Labs has recently released a new tool for Flux.1 models, which includes Redux, a model that behaves like IPAdapters and can be used to remix images and text prompts together. The Redux model is small and light, weighing in at only 123 Mb in size, making it easy to run locally on your GPU.
Getting Started
Before you start playing with Redux, make sure your ComfyUI is updated to the latest version, which includes the latest supported features. You can download the Redux model and start experimenting with it right away.
Default Workflow
You can use the default workflow provided by downloading the PNG image and loading it into your ComfyUI. This will get you started with using Redux and experimenting with it.
Tips and Tricks
I found that the text prompt does not have a significant impact on the output unless you dial down the strength of the image(s) using the Strength setting in the Apply Style Model node. If you don’t see a strength value, you may not be on the latest ComfyUI version – update.
Creating a 3-Image Blend Setup
I chained multiple image inputs in the original workflow to create a 3-image blend setup that will create the final image using the 3 inputs. I grouped them for ease of bypassing, so you can use two or one image input if you prefer.
Adding a LoRA Node
I also added a LoRA node, which lets you achieve a particular style if you wish to include a LoRA. You can bypass any nodes that you don’t want to use.
Conclusion
Redux is an exciting new tool for remixing images and text prompts, and I hope this article has given you a good introduction to getting started with it. Remember to update your ComfyUI to the latest version and experiment with different settings to achieve the desired output.
FAQs
Q: What is the size of the Redux model?
A: The Redux model is 123 Mb in size, making it easy to run locally on your GPU.
Q: How do I get started with Redux?
A: Download the Redux model and load it into your ComfyUI. You can also use the default workflow provided by downloading the PNG image and loading it into your ComfyUI.
Q: What is the impact of the text prompt on the output?
A: The text prompt does not have a significant impact on the output unless you dial down the strength of the image(s) using the Strength setting in the Apply Style Model node.
Q: How do I add a LoRA node to my workflow?
A: You can add a LoRA node to your workflow by downloading the Redux model and loading it into your ComfyUI. Then, add the LoRA node to your workflow and adjust the settings as needed.
