🌈Read Me Please on How to Use 🦄
Pony FaeTality is a versatile Pony Model that has a natural 2.5D look but can also do many other styles without having to utilize style LoRAs. 🥰
Version 1.1 Update (Chat GPT Captioned)
I made a small update focused on improving the lighting, I trained a lora on 120 images that were specifically captioned with GPT-4 to test how well GPT handles captions. Although the improvements in lighting aren't dramatically significant, there is a slight enhancement. I'll have to experiment again (whenever I have time 😿) using a much larger dataset to see more substantial results probably. 🦄
Using descriptions like this helps a lot with the effect: bright colors, high contrast, dark background, vivid lighting
Example of an image in the GPT-4 dataset along with the caption given
heart-shaped candle, vibrant melting wax, dynamic splashes, candle flame, bright colors, high contrast, dark background, vivid lighting, celebration, romance, abstract, love concept, glowing effect, sparkling particles, multicolored drips, wax flow, illuminated, artistic, creative photography, energy, melting in motion, high resolution, centerpiece, decorative, festive mood
Tips and Tricks
Always use score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up
Use Clip Skip 2
This model was trained using both blip and WD captions and tags. Utilize danbooru tags when prompting.
Certain keywords will have a bigger effect on the outputs, they are: realistic, freckles, large lips, beautiful eyes
When prompting less is better I've found, don't go crazy adding a bunch of descriptions
Go easy on the negatives, 3D in the negatives has one of the biggest effects, and this will tone down the trained style a lot if you don't like the look. If you're going to use negatives make sure they're targeted ones.
Because it's a Pony base, you can utilize the data selection tags of: 'source_pony', 'source_furry', 'source_cartoon' and 'source_anime' and ratings of 'rating_safe', 'rating_questionable' and 'rating_explicit'.
Recommended aspect ratios of 1024x1024, 832x1216, 1216x832
What I'm Good At
2.5D People, Animals, Furries
Adds pretty details and backgrounds a lot of the time naturally
NSFW
Cute Things
What I'm Bad At
Non-NSFW Complex Scenes or Objects
Sometimes I make people have giraffe necks, you can fix this by adding 'long neck' in the negatives. 🦒
Photorealism
Sometimes I like to make random things into animals 🤷♀️
How I Was Made
I started this project over a month ago by trying to finetune Faetastic SDXL to have a more 2.5D style. The entirety of the dataset is generated images. 95% of the images were my generated images of a prompt style that gave an awesome 2.5D look with FaeTasticSDXL...and I wanted a checkpoint that always had that style.
One of my images from the dataset
I was successful and the finetune is really great, but I decided I wanted it to be good at NSFW, which turned out to be pretty difficult and I had some pretty hilarious outputs.
Around this time Pony started to get really popular, because I'm busy so much with Civitai I just decided to take a shortcut and swap to Pony as the base model to achieve my goals.
So I finetuned then trained 4 different LoRAs and smashed those babies in. LoRAs were to nail down the style more + I also added my FaeTastic Details dataset into them, which seems to have given it that detailed cute vibe that I enjoy.
What I'm Going to Try to Improve (When I Can 😅)
Better Darkness/Shadows
Try to get rid of oddities
A better understanding of "normal things"
Description
FAQ
Comments (78)
1.1 New version, super excited! This model rocks my socks! Many thanks!! Edit, sadly I cannot use 1.1, its much harder to get characters to "connect" or be in the same scene. :( Not sure what broke, but 1.0 still rocks! Hopefully can fix it.
Nice
Best
Versatile is a poor description of this model as it seems highly incapable of portraying male actors. This can be seen in some example images below with female faces pasted onto otherwise male bodies. Female bias is so strong in the model that I tried with man in pos tags and female in neg tags and got 100% babes in response. Not recommended unless you are not wanting to generate any male actors.
Personally, I rarely if ever prompt for male characters and I also don't prompt for porn. I use this model for making graphics or cute animals...it's just built on pony so in the gallery porn is what everyone seems to use it for. Additionally, it can do many styles...
However, I see numerous images of male characters in the gallery, some images with several. They don't look female to me, maybe check out how those creators are prompting if you're having issues.
@jennibear Try 1boy instead of male.
I have problems getting males when I do not want them, LOL. Play with your prompting for a bit.
The quality of the faces this model generates are much higher than the base Pony model! Thank you for that!
But I have noticed a strong tendency for missing fingers - four fingered hands and where the middle and ring fingers are merged together. Don't get me wrong, hands have always been a weakness of stable diffusion, I just want to point it out in case I'm doing something wrong or if its something the author can address in a later version of this model.
The model also has a strong tendency to crop the tops of characters heads off (like 15 out of 25 images). It generates a great image, but the tops of their heads are out of frame - a lot.
Again, great model, creates beautiful faces, beautiful female characters!
I actually have never experienced that. It makes me wonder what resolution you're using to generate images
@Faeia base resolution of 768x1152. Should I be going taller or shorter?
@ttb2112 832x1216 or 1024x1024 is what I recommend, and also is the resolution of the images that I trained on.
@Faeia Thanks! I will give that a shot. I'm sure that's the issue. I see that you specified it in the model description, but I callously ignored it! 😮😂 Again, thanks for the great model. The resulting images from it are amazing!
really nice model but why did you remove it from onsite generation
Added it back to the generator.
@Faeia thanks, keep up good work by the way
why can't I use this in create anymore?
This was working well for me, but is no longer available on the onsite generator, which is a shame.
yep it's shame, been using this model a lot, but it is no longer available on site :(
Added it back to the generator.
@Faeia yAY tyvm!!!!!!
TLDR it's not clear at all how anyone is concluding this model "can't draw men". Exhibit A lmao: https://civitai.com/posts/2394574
Dear heavens, you have done an amazing job Fae.
Thank you, means a lot coming from you! 💖
Good day, i´d like to build up a Tagging Index for Trained Models. Is it possible to get a hand on the tag list for this model?
See here: https://civitai.com/articles/5106/tagging-listsindex-bring-order-into-chaos
Noob question, I see these score 9...8-up, 7-up....etc in many prompts, but what do they do or mean?
This model supports a wide array of styles and aesthetics but provides an opinionated default prompt template that allows generation of high quality samples with no negative prompt and otherwise default settings
score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, just describe what you want, tag1, tag2
(previous Pony Diffusion models used a simpler score_9 quality modifier, the longer version of V6 XL version is a training issue that was too late to correct during training, you can still use score_9 but it has a much weaker effect compared to full string. You can learn more about these tags here).
From the base pony description, you can read more there: https://civitai.com/models/257749/pony-diffusion-v6-xl
Extremely versatile.
What vae used
I love the vibrant colors of this model. This is a more distinct version of PDV6. Definitely one of my favorites.Thanks for creating!
Can somebody make an inpaint model from this. I can't merge it to an inpaint model, because of an VRAM error which I can't resolve
Latest version of Automatic1111 have Soft Inpainting option, works great. Just don't forget to increase Mask Blur to 16-20 px
@tgm ok thx, I will try
Omg... Thank you so much! I love this mix of beauty and NSFW, now it's my favorite checkpoint. Wish it would be more obedient though.
This is a fantastic checkpoint! It's just about all I use right now. One thing I've noticed with it is that, left to its own choice, it absolutely LOVES to generate blue-ish clothes. Do you happen to know why that is? Unless I specificy something else the clothes will be of a shade of blue nearly every single time.
Make someone to ride a horse without creating centaur is so hard...
Godlike! Love this one, my favourite actually, can't even use anything else after FaeTality 😍
Thank you
I've finally gotten around to giving this a proper go, and I'm really loving the look of it. Nice work!
Works well with Davant Studio, img2img and ControlNet
Just perfect
What a pity it is not a bit more realistic, but I love to use it anyway <3
That's the point of the model though. Why don't you just use something like Pony Realism if you want more realistic outputs?
Won-der-ful!
I post images made with this, yet it never shows in the gallery here. What gives and why bother?
I think the creator has hidden NSFW images by default, you have to manually select to view NSFW images - could this be what is happening to your pictures?
@jimbojones010351 Not really. The ones I wanted to post were not like that.
It seems prone to cropping off the top of the subject's head. No amount of prompting seems to address it. It either crops, or occasionally it doesn't. Very frustrating. I'll try 1.0.
Seriously though, look at all of the example images. A much higher percentage of them are close cropped to the face, losing the top of the head, regardless of prompting. I'm guessing it has a training bias for cropping closer to the subject.
Continuously encountering this problem too.
Something about this filter almost always yields great results
Has that Dreamshaper look to the generations but with the control from SDXL and Pony. This model is a brilliant work of engineering, excellent job!
This is my go to. Great job!
Also i recommend adding the 'semi realistic' and '2.5d' tags as the model does those styles very well.
Dear friends, please share some pony models that are good at generating realistic Asian women. I've been looking for a long time and haven't found any. Thank you
Have you tried?https://civitai.com/models/402800
Amazing execution
Great if you making Pixar style drawings.
It no longer works! :(
I'm sorry. It should be back on the generator next week though.
@Faeia oh good i though it faded out of existence :)
why it no longer works?
unless users put a bid on it and win the model can't be used.
As I understand it, it will not be possible to generate a picture with the help of this model?
One of the greatest set of checkpoints on Civit!
I love it this really works though I have noticed that it gets taken off from time to time as of recently.
This is my favorite Pony checkpoint. Any chance of a Illustrious version? Or anyone know a similar one for Illustrious?
plantmilk for illustrious is kinda similar https://civitai.com/models/1162518?modelVersionId=1714002
why does it show "some resources used to generate this image are unavailable"?
Sorry, it's no longer possible to create for free. It just says download. I have no idea why.
weird okay I will try and download this but what happens when I do get it downloaded what then can I still use it?
@lonewolfe777 No idea
@pinkamenausuario I see so I can download it but I can't do squat with it wow just wow.
in this page there is a "bid" system where you put yellow buzz, to have the change to get the checkpoint for a week if doesn't, make it we can use until maybe it get pass next week
@LINDAfifty seriously we'll that's one way to get buzz and money i guess.
Just started testing this - seems great
Best Pony model I've ever used
It could generate futas with male faces effortlessly
Details
Files
ponyFaetality_v11.safetensors
Mirrors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
pony123_v10.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
AICREATIONSART.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
ponyFaetality_v11.safetensors
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.