Hey everyone, Adam here!
After way too many late-night coding sprints and caffeine-fuelled prompt tests, I’m finally ready to share my first solo creation with the world. I built it because I got tired of losing track of my characters and locations every time I switched to a different scene, and I figured other AI-manga folks might be in the same boat. Would love your honest feedback and ideas for where to take it next!
The pain
• GPT-Image-1 makes gorgeous panels, but it forgets your hero’s face after one prompt
• Managing folders of refs & re-prompting kills creative flow
The fix: MangaBuilder
• Built around SOT image models for fast, on-model redraws
• *Reference images for characters & locations live *** inside the prompt workflow… re-prompt instantly without digging through folders
• Snap-together panel grids in-browser, skip Photoshop
• Unlimited image uploads, plus a free tier to storyboard a few panels and see if it clicks
Try it now → [buildmanga.com](http://buildmanga.com)
Public beta: feedback & feature requests welcome!
[noobxi](https://discuss.huggingface.co/u/noobxi)
4 Hey how does this compare to similar tools like kyukoma.com? Also seems like your website is not working