This is a very old model of mine and does not conform to my new very high-quality standard established with my "Essenz" model series, so please do not use this! It will be replaced in the near future with a new and better model in the "Essenz" series!
See here for an article about my training method to create near-perfect style and character LoRa's for SDXL 1.0!
I highly recommend using this model with an upscaler workflow. I find that it can greatly increase the already great image quality - depending on the prompt. You can find such an example workflow in the first example image of this post (the one that serves as the thumbnail of this model).
Current version: v1.0 | Recommended weight: 1.0 - or lower when you combine it with other LoRa's | Recommended VAE: SDXL 0.9 | Recommended model: SDXL 1.0 | Recommended settings: 35 steps | DPM++ 2M Karras | CFG 7 | Recommended negative prompt: text, watermark, signature, abstract, burned, instagram, low-quality, low-res, action figure, child drawing, jpeg, low contrast, glossy, puppet, cliché, stock photo, magazine cover (I have done a lot of testing with many dozens of prompt words and this is the best universal negative prompt I could find, all other common negative prompt words people use I found to have no or only a negative effect on the output)
I recommend using the official SDXL 1.0 model that has the SDXL 0.9 VAE already integrated, which you can find here.
Description
A semi-realistic concept render artstyle LoRa for SDXL 1.0, aimed at delivering high-quality 3D renders in the style of Unreal Engine game concept renders and the like.
Trigger words
"image in zz artstyle"
Changelog
v1.0
Initial release
Additional Info
Credit is not mandatory but I would still appreciate credit when posting images using these LoRa's on social media! Feel free to @ my Twitter and/or Instagram accounts too (@ai_characters)! I would love to know when it is used!
I share all of my model's and training knowledge for free, but donations to my Ko-Fi would be greatly appreciated! Testing so much and training so many model's costs a lot of money.
Feel free to also try out my RVC2 AI voice models which you can find here!
Stats
Dataset:
56x photos
Training times (for 100 epochs):
Artstyle (at 2 repeats): ~1h 30min [???s/it]
Epoch chosen for the final model:
40