<h1 id="krea2-style-gpt-image-2-lora">Krea2 Style GPT Image-2 LoRA</h1><p>A Krea2 LoRA trained on 450 images generated using GPT Image 2.</p><p>Designed to bring recurring visual and rendering characteristics of GPT Image 2 into your base model, while keeping your prompt's subject under your control.</p><h3 id="recommended-strength">Recommended Strength</h3><p><strong>0.8–1.0</strong></p><h3 id="generation-settings">Generation Settings</h3><p><strong>Samplers used:</strong> euler, euler_a, er_sde<br /><strong>Scheduler:</strong> Simple</p><h3 id="training">Training</h3><p><strong>Images:</strong> 450<br /><strong>Network Rank:</strong> 64<br /><strong>Training Steps:</strong> 5,500</p>

Files with SHA256: f68763b9094db449c90a3650e0f9f72aa3949d8d8c84ef3c4cafc51e5b6231f8