Hello ♥
for whatever reason you want to show me appreciation, you can: ❤️Ko-Fi❤️
This is my result from experimenting with Pony Models.
This model aims to create slightly more realistic images.
Pretty sure that this model still has weaknesses, but with your help I can of course improve it faster
You can create all kind of images with this Checkpoint, but if you enter nothing, for pony unknown words or an unclear prompt, it always tries to generate women.
You can see some examples in the Model Images.
Attention, the model has an urge to take nsfw pictures.
take in mind that i have informed you :D
Recommended Settings:
Pony Model special feature: (check documentary in Original Pony Model)
Sampling Method Euler A
Sampling Steps 20 or more
CFG Scale 5-7
Upscaler: SwinIR _4x (this is my all time favorite, the most others works well too )
(Hires steps: 15; Denoise: 0.28, upscale by 1.5 - 4 (depends on your GPU))
📌 USE sdxl_vae
you have to add the following trigger in
Positive Prompt
score_9, score_8_up, score_7_up,
Negative Prompt
score_6_up, score_5_up, score_4_up
Additional Trigger (optional in positive or negative prompt)
'source_pony', 'source_furry', 'source_cartoon' and 'source_anime'
I can also recommend using the these embeddings
All Images generated in SD Forge with Nvidia RTX 4090
Please test other Settings and share with the communty, so i will learn about the Settings and can optimize.
Description
FAQ
Comments (28)
Earlier today I decided I was done trying Pony checkpoints (not against them, they just don't tend to do well at all with random prompting), but I of course have to try this one.
I have the same problem, it's really something else. But I wanted to start with it.
But I'm anything but a pony pro
2x 1k Buzz for the best Image in 24 hours! ♥
One SFW (PG, PG-13, R) and one NSFW (X, XXX) Image. Both pictures will be pinned on top of this model!
i choose the Winners:
SFW Winner: @R0CKETB0Y https://civitai.com/images/12366614
NSFW Winner: @Venion https://civitai.com/images/12348841
Images got the Buzz.
Congratz and thank @all
@AIDigitalMediaAgency Thanks a bunch! Great model! ;)
@R0CKETB0Y Thank you and congratz :)
@AIDigitalMediaAgency Thank you very much the model is great! Too bad I can't exchange my picture for one of my girlfriend @Calamity82 but we will both continue to squeeze this model and see what comes out. :) (Btw. no Lora's were used in her or my pictures)
@Venion Thank you so much for the experience!
Haha, Pony? FR? Okay, I'll give it a try of course 😂
don't hate the player, hate the game :D
@AIDigitalMediaAgency 😂 Haha, I like both the player and the game. It's just a surprise.
@georgebanjog believe me, for me it's even more of a surprise. so far it's been a fight with Pony. let's see if it can turn into something.
Hi there !
Thanks, for this wonderful Checkpoint. It almost behaves, like an SDXL-Checkpoint...follows the Promt like a Dream.
While looking at the Gallery here, I was wondering about the unusual low amount of P0rn (for a Pony-Checkpoint)...do you want it like that? ...because, I am on the Way to Post some...
Let me know, if you don´t want to see P0rn in this Gallery ;-)
Cheers,
D.
Thank you very much for your kind words.
I think not many people have noticed the checkpoint yet, hence the few pictures.
Feel free to post all your results, the filters take care of everything :)
Did artist knowledge from ForRealXL make it in?
what do you mean?
@AIDigitalMediaAgency Many non Pony models have artist knowledge ("by Rembrandt", by H.R. Giger etc). Just wondering if things like that can make it into a merge with Pony? Since Pony V6 had all artist knowledge removed.
@EsheAI I think this should also be removed, but I haven't tested it yet.
this can certainly also be installed in pony models.
in this "Pony ForReal" Merge is no "ForRealXL" included
@AIDigitalMediaAgency Oh, so it just the realistic visuals added in. It's cause I think pony models would be game changer if they actually had the ability to use artists. Ah well!
How was the cover animation made with this model? This model does not seem to be compatible with AnimateDiff SDXL or HotshotXL, like other Pony based models. Cheers!
Extract every singel frame of the original video. Than "batch convert" every single picture in you comfyui worklfow to your style (you have to build one) and that merge all the images together for the new video. You will need some freeware tools to extract the frames and one to build a clip from the new frames. Its not nice and not fast....but if you dont come up with something better - voila.
@gl4mdiva I do know how to do video to video, but this looks like it's an animation made with this model specifically and I would like to know how. I have been at my wit's end looking for a solution to animating with PonyXL models! It simply does not produce temporally consistent video when using img2img without another model like AnimateDiff. Thank you though!
@markury its most likely he captured the poses with openpose to get the animation look
I’ve seen your work, man. Very, very good!!! But forgive the ignorance, what are or do the tags score_9, score_8, etc? Is there somewhere I can see/study about it?
Edit: I just learned in your article here at Civit! Thank you so much, man!
Please check the documentation of original Pony Model:
Aside from this model what's the next best model that can be used on civitai? I ask cause I only have a phone to use.
i´am not a pony expert. maybe you check the original pony model first.