ComfyUI: Iterative character consistency using generated outputs

Demo summary
The creator shows how to take a generated comic book character output and feed it back into the 'Load Image' node as a new reference to maintain style consistency across different scenes.
Step-by-step
- Locate the generated output image of your character
- Copy the image file path or name from the output node
- Paste the image path into the 'Load Image' node to use it as a new input
- Update your prompt to describe the character in a new scene or action
- Run the generation to produce a new image based on the previous output
Options
- Use subfolders for organization by using a forward slash in the filename (e.g., 'context/image_name')
Tips
- Feed generated outputs back into the input to maintain a consistent style for faces, clothing, or objects
- Use this iterative method for product shoots or creating multi-panel comic books
- Experiment with more advanced prompts once the basic feedback loop is running
Highlights
“this is just me showing you how to do the simple stuff, how to get it running”
All demos from “NEW Flux Kontext tutorial. Consistent Characters FREE & EASY!”
2:030:57Setting up Flux Kontext in ComfyUIThe creator demonstrates how to load the Flux Kontext model, configure CLIP loaders (T5 and Clip L), and refresh the model manager to ensure all nodes are active in ComfyUI.ComfyUI· AI Image Generator
3:021:04Generating a consistent character with Flux KontextThe user uploads a reference image of himself and uses a text prompt to generate a new image of himself riding a motorcycle while maintaining facial and clothing consistency using Flux Kontext.ComfyUI· AI Person Generator
4:090:55Anime style transfer and text renderingThe demonstration shows how to transform a real photo into an anime character eating in a restaurant while successfully rendering specific text in a chat bubble using the Flux Kontext model.ComfyUI· AI Style Transfer
5:191:10Iterative character consistency using generated outputsCurrentThe creator shows how to take a generated comic book character output and feed it back into the 'Load Image' node as a new reference to maintain style consistency across different scenes.ComfyUI· AI Person Generator
8:021:24Merging multiple characters with Image ConcatenateThe video demonstrates using an 'Image Concatenate' node to combine two different reference images, allowing Flux Kontext to generate a single scene featuring both characters walking together.ComfyUI· AI Image Combiner- Watch “NEW Flux Kontext tutorial. Consistent Characters FREE & EASY!” →
AI Person Generator
3:021:04Generating a consistent character with Flux KontextThe user uploads a reference image of himself and uses a text prompt to generate a new image of himself riding a motorcycle while maintaining facial and clothing consistency using Flux Kontext.Sebastian Kamph
5:191:10Iterative character consistency using generated outputsCurrentThe creator shows how to take a generated comic book character output and feed it back into the 'Load Image' node as a new reference to maintain style consistency across different scenes.Sebastian Kamph
5:033:47Generating character angles and expressions from a reference photoThe user inputs a photo of a woman into the workflow, uses an 'Image Describer' node to generate a prompt, and runs multiple generation nodes to produce profile views, back views, and various facial expressions.Ľudovít Nastišin
6:590:25Generate an AI influencer imageThe user demonstrates the 'Instagram image' preset to create a character with a specific pose and 9:16 aspect ratio.Manny ai
ComfyUI