Skip to content
This repository has been archived by the owner on Mar 11, 2023. It is now read-only.

txt2img XYZ 'Prompt S/R' iterates 'replacement word' on first pass #14

Open
NMNegative opened this issue Jan 28, 2023 · 0 comments
Open

Comments

@NMNegative
Copy link

Hi, hope this is a bug and not just me, try to explain it :) If I set say:

haircolor, black, red, blonde

in one of the txt2img XYZ script S/R fields and include (haircolor hair:1.2) in my prompt, I would expect most simply, a black hair, a redhead and a blonde in my results, but there are four passes, the first one just trying to actually prompt 'haircolor hair'.

In the img2img XYZ S/R script it seems to function correctly, and also writes the correct selected field from each S/R option to the filename / metadata, wheras the txt2image will write every filename / meta as eg: haircolor.png

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant