show what the neural network will generate.

[rudraity]

thank you for the work you have done.
Is it possible to show that the model will generate in advance, otherwise it produces the same thing with minimal changes on anime mech gestalt, anime mech medium armor a big size, anime mech medium armor a normal size. 2 dollars and an hour of time to generate and the same thing comes out.

[gn008]

Hi! One option is to use https://github.com/bytedance/MVDream to generate the multiview images, and you could check the images to see if changing prompt will make a difference. There will be differences between the multiview images and the final 3D result, but it should be close enough