I've been keeping an eye on Qwen-Edit/Wan 2.2 shenanigans and they are interesting: however actually running those types of models is too cumbersome and in the end unclear if it's actually worth it over the $0.04/image for Nano Banana.
If that's true, it seems worth getting past the 'cumbersome' aspects. This tech may not put Hollywood out of business, but it's clear that the process of filmmaking won't be recognizable in 10 years if amateurs can really do this in their basements today.
Adobe's conference last week points to the future of image gen. Visual tools where you mold images like clay. Hands on.
Comfy appeals to the 0.01% that like toolkits like TouchDesigner, Nannou, and ShaderToy.
https://www.youtube.com/watch?v=YqAAFX1XXY8 - dynamic 3D scene relighting is insane, check out the 3:45 mark.
https://www.youtube.com/watch?v=BLxFn_BFB5c - molding photos like clay in 3D is absolutely wild at the 3:58 mark.
I don't have links to everything. They presented a deluge of really smart editing tools and gave their vision for the future of media creation.
Tangible, moldable, visual, fast, and easy.