from Team Crystal Clear
You may now generate images at blazing speeds with Crystal Clear Lightning.
Pair with NVIDIA Tensor-RT for even faster generations.This model was trained with features only available in Crystal Clear XL One also known as RunDiffusion Photo
Join our DISCORD
Get in touch for business models and commissions.
Click for FEATURES list
Crystal Clear XL - Lightning
An all rounded model with extra attention to photorealism.
Trained from the ground up on modified versions of lightning and sdxl turbo, this iteration does not have a sample or cfg ceiling.
Model includes a multitude of features only available in Crystal Clear One aka RunDifussion Photo. Click for Features Preview
Generate images from 4 samples, and a CFG as low as 0.75, up to your own preference.
NOTE: High CFG values will require incremental adjustments to the number of samples, since it has a tendency to cause over exposure of the colors.
This model is free for personal use. For commercial use read below.
If you are to use this model in your merges please give us a much deserved holla.
We do not want to start flagging our content in order to find out if it's being used without our permission, so please, be a team player and let us know.
About Team Crystal Clear
We might not necessarily be known for the amount of content made public on Civitai, but we are known in the industry since we focus on commercial development of Stable Diffusion content for game developers, automakers, instagram influencers and models, stable diffusion platform providers, freelancers and other private and public companies.
Such content is considered confidential and it is at our client's discretion if they want to release it or not.If you are in need of specific Stable Diffusion content to fit your particular business needs, please visit our DISCORD for more information and quotations.
We have a team of highly skilled people with access to some of the latest hardware.
If we helped save on your business costs by using this model to generate content for commercial purposes, you may express your gratitude by clicking this DONATE button.
Description
This model does not require a VAE.
It was built from the ground up to include a VAE that will provide best output in relation to the prompt.
FAQ
Comments (5)
So happy to see this, it has been one of my favorite checkpoints since I discovered it months ago!
Can create images so realistic, some people can't tell they're AI generated
Works great with "CFG Scale" <=5
I took my first steps in AI image generation on an Australian website commonly abbreviated “NC.” When Crystal Clear Lightning was introduced along with several other SDXL Lightning checkpoints, “CCL” (as I like to call it for short) quickly became my favorite. It stood out from the others with its different aesthetic and bright colors (I later learned that the color intensity correlates with the CFG value and can be easily overcooked). For me, CCL is an all-around checkpoint that is equally suited for realism, fantasy, and artistic creations. In addition to the luminous colors and versatility, I particularly appreciate its ability to create aesthetically pleasing portraits. Some models focus on recognizability and tend toward caricature-like hyperrealism. Others are so heavily aesthetically idealized that all faces end up looking somewhat the same. In my opinion, CCL has struck just the right balance between realism/recognizability and aesthetic idealization to create beautiful, appealing portraits.
What a shame that I seem to be the only one still using this checkpoint. You can create such awesome results with it, when you find out how to play with the proper LoRAs and triggers.
For me using the DMD2 speed LoRA proved an incredibly useful step. Combine this with the Aquaart LoRA, or other style LoRAs, and the results are just "wow". I love it!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.