Have any way to improve the quality of the generated model?

[2621690255]

Hi, all!
stable-fast-3d have any parameters that can improve the quality of the generated model?
such as increasing the number of triangles used in inference.

[mattia-z]

I'm interested too, I've tried to upscale the input image to 4096x4096 and then run inference with parameter --texture-resolution=4096. But other than a quite big increase of GPU memory usage, I've not noticed any differences