Search operators for tags

[Kinomora]

Perhaps it is already possible, but I would like to see the option to use search operators to generate more "variety" of images.
Currently if I want to see a cat or a dog driving a car then I would need to do 2 different prompts.
Similarly, if I want to see an anime girl with green eyes or blue eyes, I would need two different prompts.

It would be a great quality of life improvement to be able to use simply search operands to expand what the prompt can generate.
I propose that the system randomly pick whichever option to use for the current image its working on to avoid issues with not having enough batch count/batch size to hit every variation and combination provided
Example prompts:
sedan, red car OR green car, cat OR dog, from side
girl, long hair, green eyes OR blue eyes OR brown eyes, looking at viewer

Bonus credit if this could be implemented in a way that allows multiple prompt "groups" to be created.
Example: {girl, long hair, green eyes OR blue eyes}, {boy, short hair OR long hair, brown eyes}

[ResleyZ]

This is currently possible with custom scripts.
While it won't show which options are available on their own, it will allow to use different prompts.

As an example you can use these scripts

• Advanced prompt matrix
• Wildcards
• Dynamic Prompt Templates

Personally I use the Advanced Prompt Matrix script, and when enabled, I can use these prompts
Different hair lengths;
<very short hair|short hair|long hair|very long hair|big hair>

Different hair styles:
<pixie cut|bob cut|flipped hair|hime cut|twintails|messy hair|curly hair|drill hair|pointy hair|wavy hair|spiked hair>

Combined with each other:
<pixie cut, short hair|medium hair,bob cut|flipped hair, medium hair|long hair,hime cut|twintails,long hair|very long hair,messy hair|curly hair,very long hair|very long hair,drill hair|medium hair,pointy hair|very long hair, wavy hair|medium hair,spiked hair|short hair, short hair with long locks,sidelocks>

[Kinomora]

Thank you! This looks like exactly what I'm looking for!
Does using the script "Advanced prompt matrix" prevent you from generating a >1 batch count? That seems to slightly defeat the purpose of having multiple tags in the prompts

[ResleyZ]

Using the hair tags for example, <very short hair|short hair|long hair|very long hair|big hair> this generates not 1, but 5 images, with the different prompts being a different image.

So if you would set batch count /sizet o 2, it would make 10 images per OR prompt, atleast with the Advanced Prompt Matrix plugin.
If you want to adjust batch count/size, I'd suggest to ask the repo's owner for that, or maybe have a look at the scripts .py file, and see if you can change it there.

[Kinomora]

I was actually able to get Batch Count >1 working with Dynamic Prompts instead :) It will generate multiple pictures with a random selection of the tags I used (listed below) (like what I was wanting) but the one thing I've noticed is that it seems to be generating a lot of multi-focus images- as if it sees the options and "latches on" to seed images that match more than 1 of the tags. I'm able to combat this tendency by using "solo focus" as a tag but it seems strange.

masterpiece, best quality, {small breasts|medium breasts}, thighhighs, long hair, blush, [[[abs]]], crop top, choker, sweat, shorts, {brown eyes|green eyes|blue eyes}, {white hair|brown hair|black hair}

[ResleyZ]

I only have the multiple focus happen if I forget to actually enable the script, just to be sure, you do have it selected in the scripts drop down menu?

I haven't done much with Dynamic Prompts specific yet, so it might be implemented a bit differently.
Trying the different scripts out, possibly changing some settings from them is the best way to get the best results.

Otherwise I would recommend asking about it in their own repo, maybe it's a known issue or something else weird.