Purple Grace | Style Lora - v1.0 Alt
Prompts Recomendados
amazing quality, highres, absurdres, BREAK, masterpiece, realistic, best quality, very aesthetic, 1girl, looking at viewer, solo, semi-realism, 3l3ganc3, medium breasts, depth of field
Prompts Negativos Recomendados
bad quality, low quality, ps1 screenshot, worst quality, anime screencap, loli, child, asian, bad anatomy, watermark, text, thick lips, half-closed eyes, close-up, squinting eyes, dim lighting, twin, 2 girls
Parâmetros Recomendados
samplers
steps
cfg
clip skip
resolution
other models
Dicas
The weight range for Purple Grace Lora is 0.7 to 1.3, suitable for semi-realism and ink/watercolor painting styles.
Encouraged to try the model and provide feedback for refinements.
Works well with NoobAI and other SDXL models, with NoobAI yielding a stronger effect.
Destaques da Versão
Improved Faces detailing
More stable style.
Purple Grace Lora Overview
Weight Range:
0.7 – 1.3 (semi-realism, ink/watercolor painting style)
Description:
Purple Grace emerged from a spontaneous experiment to explore what I could achieve with Lora training. The goal was to blend traditional media styles like oil painting, ink flow, and watercolor with the Elegance Niji Lora. While it was only partially successful in Version 1, I still appreciate the result. It’s not exactly what I envisioned, but it shows promise, and I have several ideas for refining it as I learn more.
I encourage you to give it a try and share your thoughts! I’d love to hear your feedback.
Compatibility:
I’ve tested Purple Grace with various models, including NoobAI, Illustrious, and SDXL. It works well for both anime-style and more 3D styles. My personal favorite pairing has been with NoobAI but I have tested it working with other SDXL family of models just having a milder impact.
Let me know if you have any requests for adjustments or additions!
Would you like any further changes or tweaks to this? Feel free to let me know if you want any specific parts to be more detailed or concise!
Detalhes do Modelo
Tipo de modelo
Modelo base
Versão do modelo
Hash do modelo
Palavras treinadas
Criador
Discussão
Por favor, faça log in para deixar um comentário.




