This version of NerdyGirlChloe was trained on 275 images generated on <a target="_blank" rel="ugc" href="http://OpenArt.AI">OpenArt.AI</a> using the DynaVision XL model. Compared to the RealCartoonXL variant, this version leans more into an illustrative, stylized aesthetic — with brighter lighting, softer shading, and highly expressive facial features.NerdyGirlChloe is a bookish blonde LoRA trained for cute, awkward, and subtly sexy content. She’s defined by her platinum blonde messy bun, black-rimmed glasses, pale blue eyes, and soft, hourglass figure with medium natural breasts. Chloe expresses a modest, flustered energy — perfect for classroom settings, cozy cardigans, candid selfies, or private photo sessions. Her body language often includes shy smiles, hand-fidgeting, or biting her lip.Use her for:<ul><li>📸 Half-body or full-body portraits with awkward posing and expressive emotion</li><li>🛏️ NSFW lingerie scenes with quiet vulnerability</li><li>🧠 Academic or cosplay roleplay sets (TA, shy superhero, sleepy reader)</li><li>☀️ Natural lighting, soft indoor shadows, and retro styled bedrooms or libraries</li></ul>Avoid: overly stylized lighting, gothic fashion, exaggerated body distortion, or non-shy expressions (unless intentionally broken for contrast).Suggested tags:platinum blonde hair, messy bun, bangs, large round glasses, pale skin, blue eyes, cute face, bookish, shy, flustered, awkward pose, lip bite, large breasts, hourglass figure, cardigan, lingerie, bookshelf, soft lighting, bedroom, studio backdrop, biting lip, flushed cheeks<h3 id="for-tos-questions-(hey-that's-depicting-a-real-person)...">For TOS questions (hey, that’s depicting a real person)…</h3>On July 18th, 2025, CivitAI informed me that there were a few reports that I was using images of real people to generate LoRAs. They checked the training datasets and verified that all my training images were AI generated. No real people. I assume the reports were coming from "Guardians" who are just looking to score some easy Buzz by reporting every realistic LoRA they see. Any LoRA I have created before July 18th has been verified to be completely AI and not a real person. Any LoRAs I create in the future will continue to be made with AI images.

NerdyGirlChloe

This LoRA version of Chloe was trained on 512 carefully curated images generated using RealCartoonXL. It captures her shy, nerdy persona across both SFW and NSFW themes, with a focus on photorealistic lighting, expressive body language, and consistent character styling. Trained with SDXL 1.0 base, it performs best on realistic or editorial-style prompts using RealCartoon-XL V7.<ul><li>Best Checkpoint: RealCartoon-XL V7, Juggernaut XL</li><li>Clip Skip: 2</li><li>Recommended Strength: 0.9 (range: 0.6 – 1.3)</li></ul>No trigger words required — just describe Chloe’s appearance and mood directly in your prompt.

Files with SHA256: 07be9953c59f6c712b60df79502676fb3f7d624722516d7f569b714910677ee3