<h1 id="ijn-takao-(azur-lane)-character-lora">IJN Takao (Azur Lane) — Character LoRA</h1>This is my IJN Takao (Azur Lane) LoRA, trained on HassakuXLIllustrious v22, with the main focus being on reproducing Takao's recognizable appearance and character design.This LoRA can produce some pretty good 2.5D realistic results without requiring any quality tags. You can keep your prompts simple and focus on the character, outfit, pose, lighting, and composition you want to achieve.Trigger word: <code>Takao</code>Base Model: <code>HassakuXLIllustrious v22</code><h3 id="recommended-generation-settings">Recommended Generation Settings</h3><ul><li>Sampler: DPM++ 2M Karras</li><li>Steps: 28–35</li><li>CFG Scale: 5–6</li><li>Resolution: 832 × 1216 (portrait) or 1024 × 1024 (square)</li><li>LoRA Strength: 0.7–0.9</li></ul>These are starting points, not strict requirements. Adjust the LoRA strength depending on how strongly you want Takao's character features to appear, and experiment with different prompting to get the style and composition you want.<h3 id="tips-for-better-results">Tips for Better Results</h3><ul><li>Include <code>Takao</code> in your prompt to activate the LoRA.</li><li>No quality tags are required. You can focus entirely on the subject and scene.</li><li>Experiment with different outfits, poses, expressions, lighting, and compositions.</li><li>If the character's appearance starts drifting, increase the LoRA strength slightly or simplify the prompt.</li><li>If the output becomes repetitive or the character features look exaggerated, lower the strength.</li><li>Use the same HassakuXLIllustrious v22 checkpoint used during training for the most consistent results.</li></ul>The LoRA is intended for 2.5D realistic character illustrations, portraits, alternate outfits, and different visual settings. You can get some really nice results without overcomplicating your prompts.If you enjoy the LoRA, consider leaving a review and sharing your generations. I'd love to see what you come up with!Have fun generating!

Files with SHA256: d91e1be58afe01739e45b43ec6eb87b0695ca7a452d3badfdcb3310eb33b1b9e