I tried out the trend myself.
However, the online services have very strict sensitivity checks, which makes it a bit tough.
I wonder how long it will take until we can freely experiment in a local environment without such restrictions.
Here’s the process I used. Feel free to try it yourself if you’re interested:
Go to LMArena https://lmarena.ai/
Click “Generate Images,” enter a prompt, and upload a base image to generate
The model changes with each attempt, so I keep trying until I get a result I like
The prompt I used was:
Please draw the character in this image as a 1/7 scale PVC figure-style concept model. Change the background to a desk with a computer monitor.