VestalWater's Illustrious Styles for Z-Image-Turbo - v2.0 | ZImageTurbo LoRA | Civitai

VestalWater's Illustrious Styles for Z-Image-Turbo - v2.0 is a LoRA model designed to enhance the output of Z-Image-Turbo by adding a more artistic and visually appealing style to the generated images. This model aims to combine the strong prompt adherence and handling of complex prompts from Z-Image-Turbo with a refined and elegant art style. The v2.0 version introduces several improvements, including a completely new dataset of extremely cherry-picked illustrious images, resulting in a far more coherent background and less hallucinatory and messy lines. Additionally, the new dataset removes the weird striations on women's skin, and it utilizes Ostris' Training Adapter v2 to better capture the concept. The model is optimized for use with Z-Image-Turbo base model, offering a balance between prompt accuracy and artistic quality. If you appreciate this work, consider supporting the creator at ko-fi.com/vestalwater. When using Z-Image-Turbo with an illustrious LoRA over Illustrious Qwen, you get the best of both worlds: Z-Image-Turbo's prompt adherence and the art style from the LoRA. Z-Image-Turbo has great prompt adherence and handles complex prompts really well, but it doesn't output images with the most flattering art style. Illustrious, on the other hand, has a great art style but struggles with prompt demands for specific subject positions and elements in the composition. This LoRA aims to utilize both strengths, adding a (subjectively speaking) flattering art style on top of Z-Image-Turbo's capabilities. All images in the showcase have ComfyUI workflows attached. For optimal results, use the following settings: Sampler: ER_NDE (Please don't use for illustrations - it's so bad for illustrations), Scheduler: None, Lora Strength: 0.8, Steps: 8, CFG: 1. The model is available for download and can be used for both personal and commercial purposes.