Core ML Converted SDXL Model:

This model was converted to Core ML for use on Apple Silicon devices. Conversion instructions can be found here.
Provide the model to an app such as Mochi Diffusion Github / Discord to generate images.
original version is only compatible with CPU & GPU option.
Resolution is the SDXL default of 1024x1024.
This model requires macOS 14.0 or later to run properly.
This model is quantized to 8-bits.
This model was converted with a vae-encoder for use with image2image.
Descriptions are posted as-is from original model source.
Not all features and/or results may be available in CoreML format.
This model does not have the unet split into chunks.
This model does not include a safety checker (for NSFW content).
This model can not be used with ControlNet.

crystalClearXL_SDXL_8-bit:

Source(s): CivitAI

Crystal Clear XL

This is an SDXL base model converted and quantized to 8-bits.

Finetuned over SDXL1.0.

Get ready to be catapulted in a world of your own creation where the only limit is your imagination, creativity and prompt skills.

Be it photorealism, 3D, semi-realistic or cartoonish, Crystal Clear XL will have no problem getting you there with ease through its use of simple prompts and highly detailed image generation capabilities.

CCXL FEATURES

higher prompt accuracy relating to camera angles (close up, long shot)
updates to skin texture (tones and colors)
improved eye correction adjustments
studio quality and detail for realistic images
clarity and sharpness enhancements
better color accuracy and vibrance
better overall scene to light accuracy (global lighting)
reduction of overall blur filter that was present in many generations

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. Our goal was to reward the stable diffusion community, thus we created a model specifically designed to be a base model for future SDXL community creations. It was created with the purpose of providing a very accurate learning base, thus it has a high compatibility with loras, hypernetworks and textual inversion content. 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.