First, let’s look at the results (all generated by directly entering prompts, with almost no modifications — in a few cases, only the positions of the speech bubbles were adjusted):
This was created using the “Storybooks” feature on fluxsd.com, powered by the Qwen-Image-2.1-Turbo-Nunchaku model.
The tool breaks down elements — such as characters, scenes, dialogue, speech bubbles, and panel layouts — into separate components, offering creators a high degree of flexibility. You can define the artistic style, select your preferred image-generation models, freely add or modify characters and scenes, and choose from a wide variety of speech bubble designs.
It also supports three distinct formats — picture books, page-based comics, and vertical-scrolling comics (webtoons) — allowing you to unleash your creativity in the style that suits you best.
If you find that character consistency is lacking during the creative process, you can generate or upload a reference image of a character; the underlying image model will then help maintain that character’s consistent appearance throughout your work.
The entire process consists of just two steps:
- Enter your prompt and click “Create.”
- Click “Fill in this episode” in the top-right corner.
In my view, fluxsd.com currently represents the highest level of production quality for AI-generated comics and picture books.
Top comments (0)