Hi There ♥️
Here's my first semi-realistic (nsfw) mature model based on PONY.
It has a good ability to handle textual inversions and LORA efficiently, providing accurate and detailed results.
You can use it with or without a score to get the best possible results.
Thank you all for your positives feedbacks, i love you guys ♥️
Recommended parameters :
Sampling method: Euler A
Sampling steps: 40 for 512px / 25 or more for 1024px
CFG scale: 4-7
Upscaler: 4x_NMKD-Siax_200k
(Hiring step: 15; Denoise: 0.3, upscaler of 1.5)
Description
In this version, you'll find more detailed and diversified backgrounds.
Better quality faces, with ever greater detail.
FAQ
Comments (8)
I try out pretty much any model that I can get my hands on. Been grateful for every single one. I tried out version 3 beta on a break at work last night and couldn't wait for my shift to end so I could use it more. I love it. I am still amateurish in my prompting but this model has created what I intended with such accuracy and consistency I swear it must be sorcery... for real though, for me this model is number 1 on following my prompts and looks great. 👍👍
Oh thank you so much for your feedback, it makes me very happy to know that you like the ckeckpoint and that it met your expectations. :)))
Software translation:
Unfortunately, I tried to create a image of the soles focus of a girl, but the barefoot soles are very blurred.
I even entered (depth of field:2) on the negative , but the checkpoint still gives me an unclear soles.
- Well, I was trying it on the V2 version before. Just downloaded the V3 beta, and the clarity is improved over V2. . Looking forward to V3 official version.
Hi,
Thanks for your feedback, I'm going to do my best to improve this point in the next version.
I'm glad to know that v3_beta already meets your expectations better than v2. ;)
@Foxaimen Software translation:
Thank you, I have only tried V3 for about ten minutes but think it should be a marked improvement.
That's what u get
This is my favorite Pony model with a male focus so far. Not only does it retain understanding of prompts and concepts without going nuts with the prompt weighting, it also generates images in half the time on my system compared to other checkpoints. What's even better is it's been giving me FANTASTIC backdrops. Thank you so much for this :)
Ohh thank you so much, your comment make me happy :)
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.