This is a NEGATIVE embed worked for keeping your prompt consistent.
It work in different type of model.
It's made because some negative waste the prompt concept.
Giving review if you try is appreciated. Show error and maybe where it waste your prompt is suggested to work in future update.
Sorry for image without data, don't know why, but civitai stripe out data where emoji is used.
Description
v1.0
FAQ
Comments (14)
Thank you again for your work! Can you help clarify the use of this embed exactly. Is this meant to replace other embeds like easynegative, kkw-*, bad-*, yam-* etc? Also, what can one expect to get from this negative? For example, should I still include things like 3D, Anime, low resolution, etc..?
it is worked for using only it as negative (see the image info, only it on negative). Mix it with other maybe get more god or bad result. I've tested with Easy Photorealism, seem work great. I suggest you to test with same prompt and see the difference with using only it and mix it with other words/embed.
Thank you, I will do more testing.
@robproctor83 thank you for testing it. Maybe if you had time, giving your impression about it is really appreciated.
@devilkkw One other thing, off topic, but do you take requests? An improved version of the 512 facial expression embeds would be greatly appreciated. Not a lot of options for photo realistic emotions. Just a suggestion. Thanks again!
@robproctor83 i made time ago something like it, but not published. Maybe i do some test and work on it do good result, expression is hard because there are really a lot of thing to work on it. If i get good resut maybe i share it, i think not in short time(some other work before).
As I wrote in my review of TI, I have the impression that it is a "just put it in for the time being" type of product, and that putting it in will dramatically improve it! I don't expect anything like that.
It's not bad, but it's not great either. Therefore, I can't recommend it, but I think it's good to try it.
Thank you for taking time and giving a review. By your review seem it tend to blurry anime model, i try to fix it in update.
Also i have to ask, because for me is not clear (image info don't help) if you have mixed it with other embed?
Sorry but some image seem don't have it in negative, and use other embedding.
@devilkkw The "Prompt matrix" compares the difference between TI and no TI, so it looks like there is no TI, but it is used (with or without TI).
On top of that, you are wondering, "Does it work?" . (the same goes for the "photo version").
@devilkkw Shall I generate some images with the metadata you are using for later?
@kurodahuga843 sorry, don't know about it. is like x/y/z script?
@kurodahuga843 if you want.
@devilkkw Slightly similar.
X/Y/Z becomes pattern A pattern B...
With Prompt matrix, only pattern A is ON, only pattern B is ON, both are OFF, both are ON, and so on.
@kurodahuga843 thank you for let me know about it.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.