Illustrious Realism Experiment (Compatible with Illustrious LoRAs)

Introduction

The Illustrious Realism Experiment (IRE) is an attempt at a realistic Illustrious model that does not sacrifice its character knowledge, prompt adherence capabilities, and compatibility with Illustrious-based LoRAs. This was mostly accomplished by block merging WAI-illustrious-SDXL v17.0 with a mix of RealVisXL v5.0, UltraEpicAi Realism v2.0, and CyberRealistic CyberIllustrious v11.0 using Hako-mikan's SuperMerger; in addition, I merged in a custom LoRA trained on a personal dataset of high quality photos.

Usage Guidelines

Summary:

• VAE: already baked in

• Prompting: use Booru tags (see below for quality tags!)

• Sampler: euler a

• Scheduler: karras

• Steps: ~30

• CFG: 2-5

• Resolution: 1024x1024, 896x1152, and 832x1216

For the best results, you should use Booru tags and place the following tags at the start of your prompt:

masterpiece, best quality, realistic, photorealistic

Including the tags below in the negative prompt may also lead to better quality:

bad quality, worst quality, sketch, flat color, lowres

The preview images used the euler a sampler with the karras scheduler at 30 steps, but feel free to experiment with other options (the DPM++ 2M SDE sampler, for example, can produce sharper images). The recommended CFG scale is 2-5, with a lower CFG tending toward more realism, in my opinion. Good resolutions include 1024x1024, 896x1152, and 832x1216.

While the model should be compatible with most Illustrious-based LoRAs, results will vary depending on the LoRA, and you may need to lower the LoRA's strength and/or the CFG scale.

The model has the default SDXL VAE baked into it.

Credits

IRE is based on block merge of WAI-illustrious-SDXL v17.0 with RealVisXL v5.0, UltraEpicAi Realism v2.0, and CyberRealistic CyberIllustrious v11.0. Free free to create new merges using this model, but please do not use them for commercial purposes.

Version History

• v4.0: Made significant changes to the block merging formula in an attempt to improve visual fidelity and realism; incorporates a custom LoRA trained on a personal dataset of around two hundred photos.

• v3.0: Reduced graininess by integrating CyberRealistic CyberIlustrious v11.0 into the merging process; incorporates a custom LoRA trained on a personal dataset of around two hundred photos.

• v2.0: Some improvements to visual fidelity by incorporating RealVisXL V5.0 and training on a higher quality dataset for the LoRA merged into the model; incorporates custom LoRAs trained on a personal dataset of around eight hundred photos.

• v1.0: Initial version based on a block merge of WAI-illustrious-SDXL v17.0 with UltraEpicAi Realism v2.0; incorporates custom LoRAs trained on a personal dataset of around six hundred photos.