Hello Everyone,
It's getting annoying to write descriptions for each model one by one. Because my settings, namely the prompt writing styles, the prompts themselves, etc., are constantly changing, the best settings always remain in my latest model, and these settings act as a piece of atom in space for those who see old models as new. That's why I will write a guidance article and give the link to this article to my models. So this will only be valid for every model when I edit this link. Everybody is happy.
You can support me on Patreon. All subscriptions, paid or free, are appreciated. Thank you.
NOTE FOR GENERATION:
If you cannot get the quality you want in the output, you can delete Trigger Words from your prompts and try again.
This is especially true for character LoRas and Trigger Words that are recognizable by the model. Not just my LoRas, but all LoRas in general. (Here, the recognizability judgment is up to you. Generally, these are words that are written properly and have a meaning.)
This is something I learned later. Some words recognized by the models can cause problems when left as they are (without the commas). For example, when the words crystal and maiden have a counterpart in the model (which is naturally much heavier in the model), as in the Crystal Maiden keyword, the output tends to shift towards that counterpart.
You have noticed this well. But at the end of the day, it is not possible to change keywords in a trained model.
(Update in 11.08.2024)
I hope you like this LoRa model. I'll wait your excellent and inspiring images send on this LoRa. Follow me to keep up to date with what I do in the future.
Settings I use
Weight: 1.0
However, do not hesitate to try different weights. :) If I have a different weight recommendation for the model, you can see it in the model version features.
Checkpoint: ponyDiffusionV6XL_v6StartWithThisOne or other Pony Variations
You can use this LoRa on Civitai only with Pony based models. However, you can download the model and use it with any model you want (of course it must be XL).
Resolution: 832x1216 / 1216x832 / 1024x1024
These are the dimensions in the features of the civitai image generator. Since XL models are trained with 1024x1024 images, you can get the best results if the total of the measurements you provide is 2048 pixels. (832 + 1216 = 2048)
Sampling Steps: 26+
I usually use 30.
CFG Scale: 7.5
I usually use 7.5. If you reduce this, you will pay less attention to the prompts and you may get the results you do not want. On the contrary, if you increase it, you may still get bad printouts as it will give more importance to the prompts. Don't hesitate to try.
Upcscale: Use what you love.
I use, 4x-UltraSharp, 4x_foolhardy_Remacri, RealESRGAN_x4plus_anime_6B. However, civitai does not support upscaling yet.
Sampling Methods: Use what you love.
I use, Eular-a, DPM++ 2M SDE, DDIM
Triggers: You can see the triggers in the area with model information. Generally, I add at least one trigger to each of my models (whether necessary or not. Please read NORE FOR GENERATON). Additionally, do not forget to review the images I have produced as examples.
NOTE: We collect all kinds of images (Cosplay, 3D, CGI, Real, Realistic, Semi-Realistic, Anime, Draw, Sketch etc.) for a character. Therefore, as the prompts show similarity, the outputs tend to shift to the image type of the relevant image. For example, as the prompt you wrote resembles the prompt of the image I used in the training, the output shifts to the style of the similar image. To prevent this, write the style you do not want to see as negative. For example, if you do not want a "cgi" style output, add a "cgi" prompt to negative.
Some Helpful Resource
First of all, before continuing the article, I will give some links and explain what they are.
Provides detailed information about PonyXL
Pony Diffusion XL v6 Innate Character Lists
An excel file prepared by @marusame. Be sure to check it out.
@BrutalPixels A cheatsheet article by @BrutalPixels. You'll definitely want to take a look.
Pony XL Tips and Tricks: Recognized Characters, Clothing, Styles... and more!
Article written by @Jayler. I gave you an excel similar to this. Those who don't want to drown in Excel can review this article.
280+ Pony Diffusion XL Recognized Clothing List - Booru Tags (SFW)
Article written by @Viola_Aiarty. I gave you an excel similar to this. Those who don't want to drown in Excel can review this article.
320+ Pony Diffusion XL Character Hairstyles, Ears, Wings and Tails - Booru Tags (SFW)
Article written by @Viola_Aiarty. I gave you an excel similar to this. Those who don't want to drown in Excel can review this article.
(Update in 24.08.2024)
Hidden characters and styles locked inside the Pony Diffusion v6 model (Deprectated)
Article written by @catto. I gave you an excel similar to this. Those who don't want to drown in Excel can review this article. But this article seems to be deprecated. The article is still available. It may be useful to you.
(Update in 17.08.2024)
Recommendations Prompts
That Can Increase Your Production Quality
Quality tags for Pony v6 and variations: (Use in Prompt section)
score_9, score_8_up, score_7_up, score_6_up, derpibooru_p_95, You may want to write the score_6_up prompt as negative. I use it this way.
Extra Quality Prompts: (Use in Prompt section)
detailed eyes, beautiful, detailed background, perfect eyes, Use it if you want anime style: (Use in Prompt section)
source_anime, very aesthetic, anime screencap, anime coloring, Use it if you want realistic style: (Use in Prompt section)
photography, realistic sun light and shadows, photorealism, UHD --ar 9:16 --style raw --stylize 700, rating_safe, cinematic, cinematic photo, close-up, portrait, orange rim lighting, atmospheric, bokeh, dynamic angle, vibrant lighting, dramatic shadows, highly detailed, realistic, high contrast, extremely detailed, photorealistic, professional, detailed skin, depth of field, film grain,Use it if you want realistic style: (Use in Negative section)
(score_3_up, score_4_up, score_5_up), sketch, drawing, cartoon, greyscale, monochrome, anime, 3d, cgi, source_cartoon, source_furry, source_pony, source_animeTry the prompts until you get the output you want. Examining images produced by other people is also a good way to learn.
These below will draw or rather (filter) your prompt and image to these specific datasets, OR you can put these in the negative.
source_pony
source_furry
source_anime
source_cartoon
rating_safe
rating_questionable
rating_explicitScore prompts, you can use them all, or actually use a selection of what group of the scored dataset to use, less means a tighter dataset but could yield better or worse results.
score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_upscore_9, score_8_up, score_7_up, score_6_upscore_9,source, rating, and score will constrain the images used to those sources or ratings, and may help in drawing out a character, such as if you know its from an anime, or cartoon.
you can think of the score as a part of the dataset that just include images with that rating up to 9, ie. if you choose a lower score you will only pull from those lower rated iamges. You can also put some of the "score's" in the negative! an example is
score_6, score_5, score_4, chromatic abberation, artifacts, ugly, bad image,For different angles:
from_below,
from_above,
from_side,
close-up,
upper body,
portrait,
contrappostoGeneral Negative prompt for Pony: (Use in Negative section)
score_6, score_5, score_4, derpibooru_p_low, lowres, bad anatomy, bad hands, text, cropped, worst quality, low quality, jpeg artifacts, signature, watermark, username, blurry, artist name, mosaic censoring, twitter username, logo, limited palette, monochrome, (Blurred poorly drawn eyes), (out of frame), (bad quality eyes), (asymmetric eyes), long neck, elongated neck, blurred eyes, undetailed eyes, ugly face, blurred, grainy, cut off, oversharpened, sign, amateur, extra limbs, dull colours, boring, lacklustre, bad art, text, abominations, more than 2 legs, more than 2 hand, fused hands, bad proportions, colorless, glitch, bad face, distorted face, messed up eyes, deformed, extra limb, extra finger, bad hands, broken finger, black and white,Or you can Negative Embeddeds.
NOTE: The words written between the negatives above may be your target point. Then you need to delete these words. For example, if you want a black and white image, you may want to delete the prompts that affect colors, such as (black and white, monochrome, limited palette, dull colours, colorless).
(Update in 11.08.2024)
If you have any suggestions to improve this article, feel free to write.
Enjoy!
Source
ChangeLog
11.08.2024 Added NOTE FOR GENERATION title and NOTE for about negative prompts.
17.08.2024 Added 3 new resourse under Some Helpful Resource title...
24.08.2024 Added 1 new resourse under Some Helpful Resource title... (Hairstyles, Ears, Wings and Tails)