I haven't overtrained this one, to make it possible to use with my DGPD model with larger variety of prompts. But it still works pretty with lots of other models. If you struggle to get grids, use my model to get you a grid in txt2img, that use ultimateSD upscale or just img2img it with higher resolution.
I'll make overtrained one later too for those who won't be able to make grids even with my model.
You can read a manual about how to use it from here.
Tags
2x2frames, 3x3frames
Additional tags
series of images of POV missionary sex
Properties
Clip skip: 2;
Resolution: 512x512 (for 2x2frames) and 768x768 (for 3x3frames). IT"S MANDATORY!
models: DGPD, perfectDeliberate, hyperHeaven, dreamShape - worked the best for me;
*Consider trying to use hires fix with x3, x4 scale and denoising str 0.4-0.6. Consistency is much higher that way.
Enjoy)
Feel free to ask any questions here in comments, or in my discord channel.
Also I am making games with AI arts. They will be free in the future, but if you want to participate in making or have an early access - you can support me on patreon)
Also my wet hair Lora was banned here, I dunno why. You can download the last version of it from my patreon for free.
Description
FAQ
Comments (16)
if it can be used for ahegao,yeahhhhhhhh
Ok so odd question, is there a gif frame lora that is more generic?
@normalfella You can use any grid Lora with additional Networks extension. Then set text weight to 0 and alpha to 0.5 - you'll get generic version. But well, the animation won't work obviously, without control net at least) And even then results will be meh)
dang i was hoping to find a generic lora that could make frames out of anything
@normalfella Ahah) That is impossible for now, i think)
@aDDont oh thats sad but i guess its just a matter of time tell someone way smarter then me figures it out
weird even using the exact prompts, settings and seeds i can not get this to come out right, any advice?
@normalfella The same model and embeddings too?
Also try making Lora weight about 0.75 and use hiresfix or sd upscaler)
@aDDont Same model, weight, and prompts. not sure what embedings are honestly?
@normalfella textual inversions, I used some (negatives) you need to download them and put them in the embeddings folder in order to use them in your prompt.
@aDDont where can i download those?
Ok ive found all but whatever BBN is, hope thats not important :)
What is BBN in the negative prompts?