Welcome to Perfection Realistic!
This is an Illustrious XL merge with a bunch of hand-selected checkpoints and LoRAs to improve the base model's realistic capabilites.
And wow, does it deliver!
Also available on mage.space.
It's focus is on creating high quality, photorealistic, but slightly exaggerated images with a focus on women.
Do not use the Normal scheduler. Use Simple or Beta!
Use euler_a, dpmpp_2m_sde or dpmpp_3m_sde as your sampler.
To use the faster low-step settings, you need to use the DMD2 LoRA. This allows you to use 8 or 16 steps and generate very fast. Remember to generate in 2 passes, either using "Hires fix" in Forge/A1111, or by creating and chaining a second sampler node in ComfyUI. A sample workflow if included in the sample images of the model. Use this as a reference.
Each sample/preview image contains the used workflow. Here's a quick article with simpler more beginner-friendly workflow. This is a recommended starting point.
Recommended Settings
Base Generation
Steps: 24
CFG-scale: 4
Sampler: DPM++ 3M SDE
Scheduler: Simple (beta upscale scheduler)
Upscale/Refine
Steps: 8
CFG-scale: 6
Sampler: DPM++ SDE (dpmpp_sde)
Scheduler: Beta
Denoise: 0.35
Description
FAQ
Comments (15)
The 7.0 version of the model is absolutely fantastic, your best model I've used, however, the 7.1 version seems quite worse (yields worse results with same prompts, some LoRAs don't work properly. e.g. Breast Size Slider [https://civitai.com/models/481016/breast-size-slider-pony-illustrious]).
But again, a big thank-you for the wonderful 7.0 version.
I agree, 7.1 is worse.
But 8.0 is here for your lunchmoney! It's the best one so far!
@6tZ Sure thing! I can't wait to try it! I'll back then with the feedback.
Ver 7 available on site isn't good to make penises or futanari content
Please try with v8.
Very impressive. It likes to place secondary subjects into the image but I'm not entirely sure if I mind that. (v8)
Interesting. I did notice a bit more people, but didn't consider it part of the model, shrugged it off as the prompts perhaps.
v8 - definitely tends to add a secondary subject
@regularhair960 Thanks for confirming. Will investigate and try to figure out what caused this.
Meanwhile: The more the merrier?
pretty trash tbh, same body, same face, can't do asians, trouble with tall/short, don't recommend
Thanks for the feedback.
yeah that is problem with Illustrious and Pony compared to the normal original SDXL since you can't use celebrities in a prompt as a base. But you can find a few random faces off the internet and use it with faceswap when you do your prompt, merge them all together, it'll make it look more unique and will look random enough it won't even look like the original person to avoid any copyright infringements.
Sometimes i'll even take a face the AI makes, then photoshop it, change its lips, eyes, head width, etc and then put it back in the faceswap, it'll look way different and unlike anything the default AI will put out without the faceswap. Its pretty fun to see the weird results.
This is actually a really great Illustrious model, it looks more realistic than any others i have so far without looking too real life, just nearly there... its all kinda off in a good way, just like the description says, bit exaggerated. Doesn't prompt as many different things as some other Illustrious but maybe i'm not doing it right.
@Hilldog Clearly you are wrong. This model is pretty trash. This has been confirmed. It absolutely cannot do asians in any way or form, nor tall/short things. Everything has to be medium height. Don't even get me started on daschhunds. Not possible.
It's not a recommended model.
Really good models, thanks!
sorry but I don't see the difference using it
is there a trigger word for it?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.