Testing Generative AI: Rearranging Elements for New Compositions
snikolov · x · 2026-08-08
A user tested a new capability of generative image models: prompting the model to rearrange elements within an image to form new compositions. Although the model doesn't literally move pieces around but rather calls upon its generative abilities to redraw, the resulting compositions are already quite interesting and novel.
More from Multimodal
- Grok Imagine Generates Insanely Realistic AI Images and Videos — XFreeze · 2026-08-08
- Running MiniMax H3 Locally on an RTX 3060: A Hands-on Test — the_frizzy1 · 2026-08-08
- Creator Uses Kling AI to Generate Cinematic Book Trailer — VoidStateKate · 2026-08-08
- Casting Myself in Titanic Using MiniMax Video Generation — casketking · 2026-08-08
- Running MiniMax H3 Fully Local on a 16GB GPU: 8-Step Video Generation & QA Lessons — Short_Regular_7191 · 2026-08-08
- Higgsfield Launches Seedance 2.5 Video Model with 33 Days of Unlimited Free Generations — eyishazyer · 2026-08-08