After creating a realism Lora for PonyXL, I decided to create one for illustrious, as it seems to be able to create an even wider variety of characters than pony, and it seems to be a lot better at prompt adherence. This Lora should be able to make realistic images in a variety of illustrious models. I recommend you check the release notes each version of the model for what it can do.
Description
Trained using civitai's training service on the same data set as last time, but with slightly different parameters. This version seems to produce better quality faces and backgrounds, though for full body shots, you may need to use DDetailer to fix faces and fingers. Seems to work best using the DPM2 A sampler at a cfg of 2.0-2.5, at around 27-30 steps. I recommend a Lora weight around 0.8, though you may need to change it depending on what sampler you are using. Seems to work on a wide variety of concepts and characters, some characters may impact the style, but not as much as the prior version. The trigger word is still "source_photo", and I recommend using the same negatives as last time. In addition, "photo,realistic,cinematic" seems to help improve the realism quality if put in the positive.