<h1 id="rydia-final-fantasy-iv-or-realistic-character-lora">Rydia — Final Fantasy IV | Realistic Character LoRA</h1>A realistic character LoRA for Rydia from Final Fantasy IV, trained for realistic image generation.The LoRA was trained specifically to capture Rydia's recognizable character appearance while allowing her to be used with different poses, compositions, outfits, and environments.<h2 id="trigger-word">Trigger Word</h2><pre><code>Rydia_(Final_Fantasy_IV)</code></pre><h3 id="recommended-lora-strength">Recommended LoRA Strength</h3><pre><code>0.8–1.0</code></pre><hr /><h1 id="prompt">Prompt</h1>Character:<pre><code>Rydia_(Final_Fantasy_IV), green hair, long wavy hair, green eyes, curvy figure, wide hips, thick thighs,</code></pre>Outfit:<pre><code>green glossy bodysuit, gold embroidery, long green gloves, green thigh-high boots, white lace trim, gold hair ornament, pink gemstone, brown beaded necklace, bridal fingerless gauntlet gloves, fingerless opera gloves, elbow-length gloves, glossy green latex, patent leather material, delicate gold filigree, leaf acanthus gold embroidery, pointed cuff design,</code></pre>The idea is to use <code>Rydia_(Final_Fantasy_IV)</code> to activate the character, then describe the pose, clothing, camera angle, lighting, environment, etc. separately.<h1 id="recommended-negative-prompt">Recommended Negative Prompt</h1>A general realistic-generation negative prompt can be used:<pre><code>low quality, blurry, distorted face, deformed hands, extra fingers, extra limbs, bad anatomy, duplicate, poorly drawn face</code></pre>The negative prompt is not required to activate the LoRA and can be adjusted according to the checkpoint/workflow being used.<hr /><h1 id="training-information">Training Information</h1>Base Model:<pre><code>novaRealityXL_ilV90</code></pre>Architecture:<pre><code>SDXL</code></pre>Training method:<pre><code>DreamBooth LoRA</code></pre>Training images:<pre><code>124</code></pre>Repeats:<pre><code>5</code></pre>Effective training images:<pre><code>620 per epoch</code></pre>Epochs:<pre><code>3</code></pre>Total training steps:<pre><code>1860</code></pre>Resolution:<pre><code>1024 × 1024</code></pre>Bucket resolution:<pre><code>512–1024</code></pre>Bucket step:<pre><code>64</code></pre>Batch size:<pre><code>1</code></pre>Gradient accumulation:<pre><code>1</code></pre>LoRA Rank / Network Dimension:<pre><code>32</code></pre>LoRA Alpha:<pre><code>16</code></pre>Optimizer:<pre><code>8-bit AdamW</code></pre>Precision:<pre><code>BF16</code></pre>Text Encoders:<pre><code>CLIP Text Encoder 1
CLIP Text Encoder 2</code></pre>U-Net LoRA modules:<pre><code>722</code></pre>Text Encoder LoRA modules:<pre><code>264</code></pre>Regularization images:<pre><code>None</code></pre>Latent caching:<pre><code>Enabled</code></pre>xformers:<pre><code>Enabled</code></pre>Hardware:<pre><code>NVIDIA GeForce RTX 4090 24GB
128GB system RAM</code></pre>Training time:<pre><code>~24 minutes</code></pre><hr /><h1 id="character-appearance">Character Appearance</h1>The LoRA is intended to reproduce Rydia's recognizable appearance, including her characteristic:<ul><li> Green hair </li><li> Long/wavy hairstyle </li><li> Green eyes </li><li> Distinctive facial appearance </li><li> Curvy body proportions </li></ul>The original character design has been adapted toward a realistic interpretation rather than reproducing the original game's exact visual style.<hr /><h1 id="tips">Tips</h1>The model was trained on <a rel="ugc" href="https://civarchive.com/models/453428/nova-reality-xl">NovaRealityv9.0</a>It can produce images using other models like <a rel="ugc" href="https://civarchive.com/models/1125067/cyberrealistic-cyberillustrious?modelVersionId=3022570">CyberIllustrious</a> or <a rel="ugc" href="https://civarchive.com/models/1490179/unnamedixl-realistic-model-illustriousxl-v20-based">UnNamedIXL</a>But the best results were on <a target="_blank" rel="ugc" href="https://civarchive.com/models/453428/nova-reality-xl">NovaRealityv9.0 </a>as it was trained with.<hr /><h1 id="notes">Notes</h1>This LoRA was trained on 124 images with 5 repeats for 3 epochs, resulting in 1,860 optimization steps.The training dataset contained varied image compositions and aspect ratios, with SDXL bucket training enabled.The model was trained with no NSFW aspects of any image. So while the checkpoint might produce results, it may not be the best or might look uncanny lacking training data.This is a character LoRA, so the trigger is intended to identify the character rather than lock the generation to a particular pose, camera angle, or environment.

Rydia (Realistic) - Final Fantasy 4 | IllustriousXL

NotesThis LoRA was trained on 124 images with 5 repeats for 3 epochs, resulting in 1,860 optimization steps.The training dataset contained varied image compositions and aspect ratios, with SDXL bucket training enabled.The model was trained with no NSFW aspects of any image. So while the checkpoint might produce results, it may not be the best or might look uncanny lacking training data.This is a character LoRA, so the trigger is intended to identify the character rather than lock the generation to a particular pose, camera angle, or environment.

Files with SHA256: 232696e68986d288dd2c55b2f048793864c983444a69bad12db645f0b7291365