Reallustrious (Photorealism for Illustrious XL)版本Alpha (ID: 1062901)

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.

描述:

This is the initial test version of the lora, trained on around 150 images on the Civitai training service. Its a very rough version as I havent yet figure out the optimal parameters, but it still can produce realistic images with the right prompting. The trigger word is "source_photo", and I recommend you put "anime,2d,cartoon" in the negatives to further improve the realism of the image. Using the DPM2 A at a cfg of around 2-2.5 sampler seems to produce the best results (though it has a lot of difficulty with backgrounds), and I recommend a LORA weight of around 0.65-0.8. The model seems to have no issues with generic concepts, but certain characters may break this version and force it to produce more cgi style results.

训练词语: source_photo