<h1 id="porforever-style-lora-trained-on-krea-2">Porforever Style LoRA trained on Krea 2</h1>For the specific workflow, please refer to the attached file. The example images use the Krea 2 Turbo Int8 Convrot model.Recommended model weight: 0.8–1.5 (If the style is not strong enough, you can increase the weight to above 1.0.)Note: Image generation with Krea 2 Turbo is highly dependent on prompt quality. If the results are not satisfactory, please try optimizing your prompt first!具体工作流参考附件，示例图使用 Krea 2 Turbo Int8 Convrot 模型推荐模型权重：0.8-1.5（如果感觉风格不够强烈可以把权重调到1.0以上）注意：Krea 2 Turbo 的图片生成非常依赖于提示词质量，如果感觉效果不够好请优先尝试优化提示词！

Files with SHA256: e86f484423edf8b5c4a5f57338454dc8992193da81de08ed173823f7b91aba5d