Welcome to Perfection Cinematic!
This is an Illustrious XL merge with a bunch of hand-selected checkpoints and LoRAs to improve the base model's realistic capabilites.
This is a special branch-off for cinematic flavor and for when you need your images to be extra juicy and well-lit!
It's focus is on creating high quality, photorealistic and cinematic 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.
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.
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.
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 (3)
@6tZ Always been a big fan of Perfection Cinematic over all the other Perfection releases, and just like your last Fucktastic revision, PC v3.0 is an excellent evolution and improvement in every regard. Improved realism, improved image quality and improved adherence for both prompt and LoRAs, and like v2.0, still very easy and enjoyable to work with and quickly get whatever result you're trying to get.
Was honestly a bit worried after seeing you only post up LCM samples, but everything still works just fine and as expected with Euler/DPM. I'll have to post something up, but I'm honestly just having far too much fun seeing what it can do.
Really glad you decided to revisit this one, as it's now my favorite of all your models. Excellent work and thank you for sharing! <3
Thanks for the kind words.
I just use LCM with the DMD2 LoRA because it's amazing and works great, and I get more samples to work with at very little / minimal quality drawback, so I don't see a good point in not doing it. I've been doing this since I found DMD2.
But I do not bake this into the model, as I want it to stand by itself, with or without the DMD additions.
Personally, I think this version is overexposed. Many images are over-bright, so I need to figure out what caused that.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.